{ "complete": true, "report": { "bucket": "vn_S_TECH_reg", "label": "Strong teacher-like, didactic delivery", "definition": { "id": "vn_S_TECH_reg", "category": "talking_style", "head": "vn_S_TECH_reg", "pole": "high", "label": "Strong teacher-like, didactic delivery", "floor": 4.0, "canonical_dimension_name": "Teacher/Didactic Style" }, "recipe_status": "corrected_taxonomy_new_search", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P01", "general": "A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.", "inline": [ "A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.", "A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.", "A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear." ], "recipe": "P01", "cfg": 1.0, "timing_syntax": "canonical", "canonical_labels": true, "lexical_proxy": false, "origin": "new_canonical_manual_extension_requires_new_evidence" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.6898437598642937, "target_raw": 2.5868027329444887, "ce": 6.169354486465454, "genuineness": 4.139885020256043, "identity": 0.540902829170227, "wer": 0.01388888888888889, "valid_fraction": 0.95, "identity_delta": 0.006773810833692551, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.7718750137641308, "target_raw": 2.6867114067077638, "ce": 6.130832505226135, "genuineness": 4.091402506828308, "identity": 0.5395275890827179, "wer": 0.021052631578947368, "valid_fraction": 0.95, "identity_delta": 0.0053985707461833956, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.6898437598642937, "target_raw": 2.5868027329444887, "ce": 6.169354486465454, "genuineness": 4.139885020256043, "identity": 0.540902829170227, "wer": 0.01388888888888889, "valid_fraction": 0.95, "identity_delta": 0.006773810833692551, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.6640624931179346, "target_raw": 2.555402845144272, "ce": 6.179481363296508, "genuineness": 4.204214382171631, "identity": 0.6171471290290356, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.08301811069250106, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.635156218113097, "target_raw": 2.5201969027519224, "ce": 6.0150681138038635, "genuineness": 4.165302896499634, "identity": 0.5059815563261509, "wer": 0.02291344169246646, "valid_fraction": 0.9, "identity_delta": -0.028147462010383605, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.616406281122229, "target_raw": 2.4973607122898103, "ce": 6.1715387344360355, "genuineness": 4.247205269336701, "identity": 0.6036560744047165, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.06952705606818199, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.5849609362956385, "target_raw": 2.4590623497962953, "ce": 6.113557195663452, "genuineness": 4.141767823696137, "identity": 0.5509652480483055, "wer": 0.011228070175438596, "valid_fraction": 0.95, "identity_delta": 0.01683622971177101, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.5195312600172286, "target_raw": 2.3793733060359954, "ce": 6.190998291969299, "genuineness": 4.210804212093353, "identity": 0.5671309560537339, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03300193771719932, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.5007812251480972, "target_raw": 2.3565369963645937, "ce": 6.213237738609314, "genuineness": 4.287685739994049, "identity": 0.5903614863753319, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.056232468038797376, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.5642578091927852, "target_raw": 2.43384730219841, "ce": 6.103827476501465, "genuineness": 4.149298989772797, "identity": 0.4498014732729644, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.08432754506357014, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 1.7718750137641308, "target_raw": 2.6867114067077638, "ce": 6.130832505226135, "genuineness": 4.091402506828308, "identity": 0.5395275890827179, "wer": 0.021052631578947368, "valid_fraction": 0.95, "identity_delta": 0.0053985707461833956, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.6632812581820111, "target_raw": 2.5544513523578645, "ce": 6.1937782049179075, "genuineness": 4.155575060844422, "identity": 0.5552162274718284, "wer": 0.008333333333333333, "valid_fraction": 0.95, "identity_delta": 0.02108720913529396, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.6527343467452982, "target_raw": 2.541605907678604, "ce": 6.035321879386902, "genuineness": 4.062375903129578, "identity": 0.5008247524499894, "wer": 0.008187134502923975, "valid_fraction": 1, "identity_delta": -0.033304265886545184, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.6378906284792665, "target_raw": 2.5235272347927094, "ce": 6.1782899618148805, "genuineness": 4.204528200626373, "identity": 0.5668771799653769, "wer": 0.014791666666666667, "valid_fraction": 0.95, "identity_delta": 0.03274816162884235, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.6375000061173914, "target_raw": 2.5230514824390413, "ce": 6.064362311363221, "genuineness": 4.087793791294098, "identity": 0.5705316796898842, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.036402661353349686, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.6296874561077161, "target_raw": 2.513536310195923, "ce": 6.211649227142334, "genuineness": 4.212373232841491, "identity": 0.5900341719388962, "wer": 0.021637426900584796, "valid_fraction": 0.95, "identity_delta": 0.05590515360236168, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 1.6246093458276893, "target_raw": 2.507351505756378, "ce": 6.157837605476379, "genuineness": 4.15149564743042, "identity": 0.5555893331766129, "wer": 0.006066176470588235, "valid_fraction": 1, "identity_delta": 0.021460314840078355, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.6238281353613317, "target_raw": 2.506400042772293, "ce": 6.207280778884888, "genuineness": 4.204842007160186, "identity": 0.5866205334663391, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.052491515129804614, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.6164062370770105, "target_raw": 2.49736065864563, "ce": 6.2025151491165165, "genuineness": 4.221787309646606, "identity": 0.5592874508351088, "wer": 0.005409356725146199, "valid_fraction": 1, "identity_delta": 0.025158432498574256, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 1.6140625077996742, "target_raw": 2.494506150484085, "ce": 6.107203078269959, "genuineness": 4.1863276839256285, "identity": 0.5419998412951827, "wer": 0.04160818713450292, "valid_fraction": 0.9, "identity_delta": 0.007870822958648205, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.6132812630759243, "target_raw": 2.4935546457767486, "ce": 6.208670687675476, "genuineness": 4.241556799411773, "identity": 0.5807249069213867, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04659588858485222, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.613281253288098, "target_raw": 2.49355463385582, "ce": 6.184246969223023, "genuineness": 4.286116766929626, "identity": 0.5624050796031952, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": 0.02827606126666069, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.6087890471874045, "target_raw": 2.4880834221839905, "ce": 6.177495765686035, "genuineness": 4.257560658454895, "identity": 0.5784653387963772, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04433632045984268, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.6070312514528804, "target_raw": 2.485942542552948, "ce": 6.200529456138611, "genuineness": 4.2126870393753055, "identity": 0.5625344835221767, "wer": 0.00804093567251462, "valid_fraction": 1, "identity_delta": 0.028405465185642242, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.60507811517394, "target_raw": 2.4835637509822845, "ce": 6.128549003601075, "genuineness": 4.132039940357208, "identity": 0.5748856097459794, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": 0.040756591409444806, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 1.5960937665934243, "target_raw": 2.472621405124664, "ce": 6.150887751579285, "genuineness": 4.2283770680427555, "identity": 0.6095373868942261, "wer": 0.011948529411764707, "valid_fraction": 0.95, "identity_delta": 0.07540836855769158, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.5874999865417387, "target_raw": 2.462154746055603, "ce": 6.1812684535980225, "genuineness": 4.167499566078186, "identity": 0.5888942316174507, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": 0.054765213280916214, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.5859375019881523, "target_raw": 2.4602517426013946, "ce": 6.126662611961365, "genuineness": 4.222414898872375, "identity": 0.5958898693323136, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.061760850995779035, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 1.5855468551567113, "target_raw": 2.459775960445404, "ce": 6.171935868263245, "genuineness": 4.211745607852936, "identity": 0.5905861616134643, "wer": 0.011601307189542484, "valid_fraction": 0.95, "identity_delta": 0.056457143276929855, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.5851562376887496, "target_raw": 2.4593002140522002, "ce": 6.163596034049988, "genuineness": 4.241243040561676, "identity": 0.5967848032712937, "wer": 0.015935672514619883, "valid_fraction": 0.95, "identity_delta": 0.06265578493475914, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.5583984395072692, "target_raw": 2.4267109751701357, "ce": 6.135796689987183, "genuineness": 4.14239535331726, "identity": 0.5545736804604531, "wer": 0.01789624183006536, "valid_fraction": 1, "identity_delta": 0.020444662123918534, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.549609372744212, "target_raw": 2.4160064697265624, "ce": 6.169553017616272, "genuineness": 4.216452670097351, "identity": 0.5733404263854027, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.03921140804886818, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 1.5472656141033965, "target_raw": 2.413151925802231, "ce": 6.015068078041077, "genuineness": 4.169382333755493, "identity": 0.5265508010983467, "wer": 0.05526315789473684, "valid_fraction": 0.95, "identity_delta": -0.00757821723818779, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.4179687412062498, "target_raw": 2.2556768357753754, "ce": 5.893346619606018, "genuineness": 4.141453993320465, "identity": 0.5483547911047936, "wer": 0.060982112143102854, "valid_fraction": 0.95, "identity_delta": 0.014225772768259048, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.5316406070683963, "target_raw": 2.394121694564819, "ce": 6.190204048156739, "genuineness": 4.2723094582557675, "identity": 0.5341290183365345, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.5, "n": 20, "target_z": 1.5148437378416846, "target_raw": 2.373664212226868, "ce": 6.195366740226746, "genuineness": 4.239360201358795, "identity": 0.5525770708918571, "wer": 0.018421052631578946, "valid_fraction": 0.95, "identity_delta": 0.01844805255532265, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.4982421602202574, "target_raw": 2.3534445822238923, "ce": 6.19973521232605, "genuineness": 4.269171404838562, "identity": 0.5328400149941445, "wer": 0.011455108359133128, "valid_fraction": 0.95, "identity_delta": -0.0012890033423900605, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.5070312171954883, "target_raw": 2.3641490757465364, "ce": 6.188615465164185, "genuineness": 4.251912260055542, "identity": 0.514016917720437, "wer": 0.027777777777777776, "valid_fraction": 0.95, "identity_delta": -0.02011210061609745, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.6541406254205706, "target_raw": 2.5433186626434328, "ce": 6.239766292572021, "genuineness": 4.251912288665771, "identity": 0.5765025797486305, "wer": 0.007556960583276372, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 1.7628125044962828, "target_raw": 2.6756738662719726, "ce": 6.2589875507354735, "genuineness": 4.142207112312317, "identity": 0.572125246822834, "wer": 0.012138647012788889, "valid_fraction": 0.98, "identity_delta": -0.004377332925796509, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 1.7932812620971417, "target_raw": 2.7127828097343443, "ce": 6.2575578784942625, "genuineness": 4.168315415382385, "identity": 0.5612384041026235, "wer": 0.00872362724994304, "valid_fraction": 1, "identity_delta": -0.015264175646007062, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 1.8271874788644125, "target_raw": 2.7540783524513244, "ce": 6.249932889938354, "genuineness": 4.123881039619445, "identity": 0.5622481724619866, "wer": 0.005647186147186147, "valid_fraction": 1, "identity_delta": -0.014254407286643981, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": -0.03390621676727083, "ci95_z": [ -0.07578120801940103, 0.007968774484859352 ], "language_deltas": { "en": -0.02734372193646656, "de": 0.025781261852445926, "fr": 0.007812535327935733, "nl": -0.08124994066130278, "es": -0.0945312184189665 }, "languages_improved": 2, "clear_gain": false, "delta_ce": 0.007624988555908203, "delta_genuineness": 0.04443437576293945, "delta_wer": 0.003076441102756892, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "adapter_vs_best_manual": { "n": 50, "delta_z": 0.03046875760085891, "ci95_z": [ -0.03968750015293475, 0.08421875674442411 ], "language_deltas": { "en": 0.11015626949918535, "de": 0.06953126980505482, "fr": -0.09765626582875035, "nl": 0.025000026916522478, "es": 0.04531248761228226 }, "languages_improved": 4, "clear_gain": false, "delta_ce": -0.0014296722412109374, "delta_genuineness": 0.02610830307006836, "delta_wer": -0.0034150197628458495, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.10867187907571205, "ci95_z": [ 0.01656251342767427, 0.2007812447237498 ], "language_deltas": { "en": 0.10468747813032549, "de": 0.006250026304783329, "fr": 0.2500000244695658, "nl": 0.19960934827464596, "es": -0.017187481800760396 }, "languages_improved": 4, "clear_gain": true, "delta_ce": 0.01922125816345215, "delta_genuineness": -0.10970517635345459, "delta_wer": 0.0045816864295125164, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 2.5, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793046136.63897, "example": { "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "language": "es", "seed": 1701, "scale": 0.0, "cfg": 2.5, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "0097d5748ba9575ae730d61377da80c5f9c9eff65e2175166b5dd0d4b7df5dab", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8834fd58c4374688fbfadb64f13a3bdb90d720a23ec5e71441870a32f4b7ea7", "generation_seconds_batch": 23.550592538085766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.1894097328186035, "genuineness": 3.8044304847717285, "identity": 0.3826807737350464, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0097d5748ba9575ae730d61377da80c5f9c9eff65e2175166b5dd0d4b7df5dab.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "00a338c9b954f9f69dbdb550216336d63a105fed7ae351d121eab3e0c186e925", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e499083640a015a3a1a1d64145f69757229dba380ab3bfef4b25cd331610b0a", "generation_seconds_batch": 16.95474880503025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.165581703186035, "genuineness": 4.093128204345703, "identity": 0.7629072666168213, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/00a338c9b954f9f69dbdb550216336d63a105fed7ae351d121eab3e0c186e925.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "00bfa3d3ce39e9c9a6ff86d98cc6f627dcefe51fd937f712b5e139df7611f8a2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f18fc68196d8de4e685755cb73dbd105999eb2318768380586901d90f3975e91", "generation_seconds_batch": 16.34132209303789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.936007261276245, "target_z": 1.9765624625309772, "ce": 6.41180419921875, "genuineness": 4.105680465698242, "identity": 0.5269464254379272, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/00bfa3d3ce39e9c9a6ff86d98cc6f627dcefe51fd937f712b5e139df7611f8a2.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "00c0bd78ecc66823a85561dec2a15c518cf083c83d97451066416250df043df2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34caef0654ad65d6e5408e782a850009783747d2094b547caaf7d189b41c2754", "generation_seconds_batch": 24.189882898004726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.1894097328186035, "genuineness": 4.494794845581055, "identity": 0.7625499963760376, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/00c0bd78ecc66823a85561dec2a15c518cf083c83d97451066416250df043df2.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "00cf03ac9a472c95552339a883b4b990ff7a70cf59782dec7d7a409b83be0bb4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4e3553bb86d6a259f216c02d4212425f653041a94baf2416a9afb8d4fda52df", "generation_seconds_batch": 22.39065151894465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.141753673553467, "genuineness": 3.7165660858154297, "identity": 0.28774571418762207, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/00cf03ac9a472c95552339a883b4b990ff7a70cf59782dec7d7a409b83be0bb4.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "010c95df9d2140f8a23f2d611c90e65767202f42508aef9399544c77c1e6617c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25127eb0d7aff24afad41b35302224446dc7a59ba60fa2f0e190577e307d3e74", "generation_seconds_batch": 22.930848905001767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 5.903473377227783, "genuineness": 4.193545341491699, "identity": 0.6717625856399536, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/010c95df9d2140f8a23f2d611c90e65767202f42508aef9399544c77c1e6617c.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "01b058af50bd68f4a23f3a50f11b8fb531d0c9feaee20e66159479a03ef3bd2e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c45fcdafe1cac6552e5a05b2545afb977082c806c30c6bf2d6bd1541743f7dc", "generation_seconds_batch": 23.474074197001755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.135824680328369, "target_z": 2.1406250290576097, "ce": 5.8875885009765625, "genuineness": 3.898571252822876, "identity": 0.6577602624893188, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/01b058af50bd68f4a23f3a50f11b8fb531d0c9feaee20e66159479a03ef3bd2e.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "01b7c145a0bdcc53dc375c720f55832ad94669d236d6776271000754985d6012", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cdea2371e9692341ad09cf488052b39dfd899f8d516871c307313f4ca015fa34", "generation_seconds_batch": 15.29709417000413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.070269584655762, "genuineness": 3.9927117824554443, "identity": 0.2748009264469147, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/01b7c145a0bdcc53dc375c720f55832ad94669d236d6776271000754985d6012.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cf421628b2eec15b36105f988cd4b8fa1ad8cfd850d9c9771b81303f25add26", "generation_seconds_batch": 16.625318131991662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.141753673553467, "genuineness": 4.413206577301025, "identity": 0.6852396726608276, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et verifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "025a7e0cc11ab914407043ab4298976083c7f73a3f0da16c75712bccd932b9ad", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5893a18df517624230299a418d5ccdc010ed0e01babdc2ae82fc0807670da2e", "generation_seconds_batch": 22.805018635001034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 6.24500846862793, "genuineness": 4.325342178344727, "identity": 0.6960572600364685, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/025a7e0cc11ab914407043ab4298976083c7f73a3f0da16c75712bccd932b9ad.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "02850f93c171f2c5ecb3db5a0484761d7c3f0ba2cbe811d96768c2c014a5ac06", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4600b24b783a917f75b57f565af8c886c16d2bdda8c8b897e2b06a77f5ce7d20", "generation_seconds_batch": 21.20651698601432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3451571464538574, "target_z": 2.3125000428217404, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.6987532377243042, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/02850f93c171f2c5ecb3db5a0484761d7c3f0ba2cbe811d96768c2c014a5ac06.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "041121b17c149a884e71f01eff4d82163caf057ff892e361deb1eac2213e0a7c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f093c0b252ea7caac5c0b08ed81019b2cd352eb2551367541c82831fe15e8265", "generation_seconds_batch": 17.544126479071565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.268836498260498, "genuineness": 4.425758361816406, "identity": 0.6475314497947693, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/041121b17c149a884e71f01eff4d82163caf057ff892e361deb1eac2213e0a7c.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "0450f5fed9f64cad622ead055718c1adc25b7c9ee5351c9a76813971986352e2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "add2e8c9927b9c819d0708a0c76c4ef2a6539204a4602e87186dcec0b72d5a8e", "generation_seconds_batch": 18.15141965902876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.141753673553467, "genuineness": 4.118232727050781, "identity": 0.6367465853691101, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0450f5fed9f64cad622ead055718c1adc25b7c9ee5351c9a76813971986352e2.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "04556bb145122005c82482e035ea192cf0547f9e98ec6160f1eba5828dee15b8", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03e74e5a0d8a4cc4d9e1678ce1369488555a9aab573b88e8e4ba2c43500c9046", "generation_seconds_batch": 17.20948301802855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5658313035964966, "target_z": 0.8515625237048919, "ce": 6.38797664642334, "genuineness": 4.469690799713135, "identity": 0.5522128343582153, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/04556bb145122005c82482e035ea192cf0547f9e98ec6160f1eba5828dee15b8.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "045623d77334843832fc5ec303b6255a94024f786b7049d581dfd758d9b67430", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7e2f346b5c248d07a62fd39a365f2f66e1544a63af271485927d3cfc1dfa2f8", "generation_seconds_batch": 22.52482335397508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.536372661590576, "target_z": 1.6484375252342398, "ce": 6.149696350097656, "genuineness": 4.061748504638672, "identity": 0.6461885571479797, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/045623d77334843832fc5ec303b6255a94024f786b7049d581dfd758d9b67430.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "04623064b7ef137a0458259c59359f9a72bb1ec924fa3717fe01b70ef17ee031", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "073ef291ece6bec6384fbda61572ae8d8b8a00a84ed1ecaea642383843830080", "generation_seconds_batch": 16.229206256917678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0415868759155273, "target_z": 1.2421875282929355, "ce": 6.356205940246582, "genuineness": 4.319066047668457, "identity": 0.4073657989501953, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/04623064b7ef137a0458259c59359f9a72bb1ec924fa3717fe01b70ef17ee031.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "04926ab99063caa05a971f023e4cebfb96e57f81814f77522cc007b3c159e7fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "72b0ab8de8a64c4c1d737b2741ed9a81b4e0201a67e87a297261886243c50955", "generation_seconds_batch": 22.1711235078983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.221180438995361, "genuineness": 3.8044304847717285, "identity": 0.34005260467529297, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/04926ab99063caa05a971f023e4cebfb96e57f81814f77522cc007b3c159e7fa.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "0499f407c51cc2e70ba57652d80ef52c676c8e73611c73b16d0474ca514d3be8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "862f1f57601d75fe5683489a9ca9152d328c6c6a585b5e80373812e0e2fb9056", "generation_seconds_batch": 25.51077109691687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 5.685050010681152, "genuineness": 4.1998209953308105, "identity": 0.6189978122711182, "hypothesis": "J'ai trouvé la petite boîte , elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0499f407c51cc2e70ba57652d80ef52c676c8e73611c73b16d0474ca514d3be8.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "054d07faba304a9f8df1a1253153a221ede77b60593d4d240f1d0dadd8a53b76", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48e5aa5c49861e3a1dcf78a809d836dd2b14af2fbf7066057820d45d45eb1296", "generation_seconds_batch": 20.822211446007714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.324435234069824, "genuineness": 4.20609712600708, "identity": 0.32771408557891846, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/054d07faba304a9f8df1a1253153a221ede77b60593d4d240f1d0dadd8a53b76.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "0569edd57e71091a5b4dfd1fc562be293ccf031c64cd9c3aface132529abf46f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "117b52696963cf8b0bdd86d0d5aed8f58cfbccafec7195adfd5a93a4beeef85c", "generation_seconds_batch": 18.84678244800307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.252951145172119, "genuineness": 4.093128204345703, "identity": 0.6326332688331604, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0569edd57e71091a5b4dfd1fc562be293ccf031c64cd9c3aface132529abf46f.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c3f366dacdf979761948d781245e7cb329e55a8495c396bb6fb7a92d364a4a4", "generation_seconds_batch": 17.0768135729013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.2767791748046875, "genuineness": 4.582659721374512, "identity": 0.5530304908752441, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg, sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "05ab3df4a66c8c623056933a2429ef43cde7e72d5408238d177c763044d72677", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf059ea4bfbd94baedcfb1c3d05c3788375446a587771777cb401b7822a117b4", "generation_seconds_batch": 20.91654931695666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.181467056274414, "genuineness": 4.293961524963379, "identity": 0.7854712009429932, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/05ab3df4a66c8c623056933a2429ef43cde7e72d5408238d177c763044d72677.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "064fab4e7dff22b8545b68f580bdc442d9fd34078043097c80935c572bbef934", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "27ff78a350429ed24777ecd1d3dc4fadc0e4edc7f5bf1d3c2b28fd8a16ef6ccf", "generation_seconds_batch": 17.013645151047967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 5.728734970092773, "genuineness": 4.268857479095459, "identity": 0.5219371318817139, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek en daarna naast het raam. Neem ze mee als ze", "wer": 0.22727272727272727, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/064fab4e7dff22b8545b68f580bdc442d9fd34078043097c80935c572bbef934.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "069814bc115331481ca54d59d723c2dda479aaa4d6c1db3f4d1904f6487b8394", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b071db681ef627e452c66a4fecb595b456fdb26cde30b67633d2c9e17dc8934c", "generation_seconds_batch": 21.652251077000983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9473490715026855, "target_z": 0.3437500275282616, "ce": 3.465073585510254, "genuineness": 4.042920112609863, "identity": 0.14869335293769836, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/069814bc115331481ca54d59d723c2dda479aaa4d6c1db3f4d1904f6487b8394.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "06b63164c5ce1ca1bb7f02e85daae98639901cf2db18b82ea7021f2e7eb14433", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d94869889611f5396170fc6a7f3a08b0fd4d2a8be8a3f301528f6c8386149406", "generation_seconds_batch": 17.33814301295206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.897946834564209, "target_z": 1.945312477824456, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.701590895652771, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/06b63164c5ce1ca1bb7f02e85daae98639901cf2db18b82ea7021f2e7eb14433.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "06e858cdf0af49aab8bc457a3d5322034072abb7b6b078fb5388d4e507e209e4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5295f092e0fcac11c79c0769dac9cea94f01dc0bb2aa83b18d028740f0d66997", "generation_seconds_batch": 24.574281575973146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.536372661590576, "target_z": 1.6484375252342398, "ce": 6.205295085906982, "genuineness": 3.9425034523010254, "identity": 0.239877849817276, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/06e858cdf0af49aab8bc457a3d5322034072abb7b6b078fb5388d4e507e209e4.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "06fb2274464dfd8c3c72b019aebeae81cb3bb4df46f83d16f424729f2ba77cf5", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce6aa691a7ceea519a79a599f79696959280fec0b8e47a2d3ca31ec024cbdc82", "generation_seconds_batch": 14.738098341040313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.7241963148117065, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/06fb2274464dfd8c3c72b019aebeae81cb3bb4df46f83d16f424729f2ba77cf5.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "072eba7fe9538097cdfef927a9fb63da29dd85392885fb20c80be281259fe446", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9441b245c00e32903456aa62a3facc482f1114ad02277e71e53c73f1601f94f", "generation_seconds_batch": 21.13964100810699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.2767791748046875, "genuineness": 4.20609712600708, "identity": 0.5307433605194092, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/072eba7fe9538097cdfef927a9fb63da29dd85392885fb20c80be281259fe446.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "0774da218d53f967cd6652674c49e77c8ece68dff25809c3bae5108d9ae214fd", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5a419a7fca6234a58806cf7b0f2ecf2f995ddf8fb6470f6fd789b79eac76450", "generation_seconds_batch": 15.2062165789539, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.022613525390625, "genuineness": 4.18099308013916, "identity": 0.6201759576797485, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0774da218d53f967cd6652674c49e77c8ece68dff25809c3bae5108d9ae214fd.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "07dcdf389ac68cf303b44b4ded8b6d8a262e1dccd4fa7c3073cce9923065425a", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae9385f82ef54fa15a10407027a0a99d6b4ced6f50394fb5057f11d758760424", "generation_seconds_batch": 16.949406045954674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.332377910614014, "genuineness": 3.7291181087493896, "identity": 0.4299854636192322, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/07dcdf389ac68cf303b44b4ded8b6d8a262e1dccd4fa7c3073cce9923065425a.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "08fbaa3985ff6ac1ea54c1a83c2d9a809f01e73c22e09aea30fd65f0e46d4a0e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f36fbde866238b606e0a61bfe525be984ae7f8d9aa1a4f28a0ab39f3f5feeaf7", "generation_seconds_batch": 17.065669848001562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.440308094024658, "target_z": 2.39062490670978, "ce": 6.141753673553467, "genuineness": 4.256305694580078, "identity": 0.7178459167480469, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine, houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/08fbaa3985ff6ac1ea54c1a83c2d9a809f01e73c22e09aea30fd65f0e46d4a0e.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "0971f1ced4084f63ffebb72e1b9fe0696266b8f3b02ad6fb5c8d8a7f4e018f32", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bfee55ad268d64858f2ec6488c5e6bc0bf11c0604e39ac90cc49fcb1b1049eb", "generation_seconds_batch": 23.802937470958568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.2767791748046875, "genuineness": 3.986435890197754, "identity": 0.09960557520389557, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0971f1ced4084f63ffebb72e1b9fe0696266b8f3b02ad6fb5c8d8a7f4e018f32.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "09d8466453b0a5dd20275a2a4e3c6d2c90196f72747aed8c61f92e5d63e2dbe7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85679d5328737201d10934386e5a76b82c8203d44b745f1e03df6ef83e2bb374", "generation_seconds_batch": 16.80598330590874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 5.967014789581299, "genuineness": 4.231201171875, "identity": 0.46641597151756287, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/09d8466453b0a5dd20275a2a4e3c6d2c90196f72747aed8c61f92e5d63e2dbe7.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "0a1153235ce5de801133dde594f99cfe62ee836f8b7d27446fd89ad5bf39d5bb", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7ca8afd6839a2ddb334796e6bce5136fec6080c1a9435a4f9bb2380304f862ca", "generation_seconds_batch": 22.78242627205327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.260893821716309, "genuineness": 4.293961524963379, "identity": 0.5236197710037231, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0a1153235ce5de801133dde594f99cfe62ee836f8b7d27446fd89ad5bf39d5bb.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "0a258f5b265c6dfa1fbfba75608802d89bc64b936a31c5066182c3d6439addd5", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9eafe6aae9f28823266e88ce7ddbf86414506c96faf1aa92b4d01a622c43077", "generation_seconds_batch": 17.065669848001562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.070269584655762, "genuineness": 4.2751336097717285, "identity": 0.6588501930236816, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0a258f5b265c6dfa1fbfba75608802d89bc64b936a31c5066182c3d6439addd5.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "0a2c9291bb9f83dd0309c4d5b8c23ede7fd6407563ee6e5b12496acc92cfd441", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3f721ade33670aa8810a7f6144b31c995acadae91b91f5603923f7eeb6de573b", "generation_seconds_batch": 23.582460697041824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5554027557373047, "target_z": 1.6640624197092369, "ce": 5.919358730316162, "genuineness": 4.394378662109375, "identity": 0.48780643939971924, "hypothesis": "J'ai trouvé à la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0a2c9291bb9f83dd0309c4d5b8c23ede7fd6407563ee6e5b12496acc92cfd441.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "0a56a928e2057695f10778e9f51e6b248509c33d3a0648106ee8e4999e1dba92", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01f50de69667fe5ae0027b2dc5512a3dd6a3941dab87dce6d0ea18ee5f5e47bf", "generation_seconds_batch": 22.94669146090746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.1020402908325195, "genuineness": 4.319066047668457, "identity": 0.80256187915802, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0a56a928e2057695f10778e9f51e6b248509c33d3a0648106ee8e4999e1dba92.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0a8a4ad14ba47713c0d8cba70e6c859f302c27ea785abab15f35446a302e1178", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17a36f96a2870794708afbce21dc87cdaf1c49b18299c70cca435af24a179ce7", "generation_seconds_batch": 18.089015537058003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.403861999511719, "genuineness": 4.319066047668457, "identity": 0.3890339732170105, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0a8a4ad14ba47713c0d8cba70e6c859f302c27ea785abab15f35446a302e1178.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "0ac1a70df42cc6779fdc9d2a19218b1d924baf06780efd1a21988a8da738328e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b978dc5555b28159054a3a02ef53ed66eb2d3227cd33381df035313d3bb4b64", "generation_seconds_batch": 23.199069195077755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.221180438995361, "genuineness": 4.231201171875, "identity": 0.4191907048225403, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ac1a70df42cc6779fdc9d2a19218b1d924baf06780efd1a21988a8da738328e.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "0ac3361286d745495f49ee22358d912452efba5ad042a99a93459704eba5af75", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93be96cc027f2580019699eb0ed0e8941913e6da65d0f91f4fbb0270f34e23b0", "generation_seconds_batch": 16.52902959403582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.47866758704185486, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ac3361286d745495f49ee22358d912452efba5ad042a99a93459704eba5af75.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "0b0009b1d4e9b816b66663fbd9eeca50296a511fee445ab444d6aed07f7cb70b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ce3bd2b6f68cbb1ed0f350412060d406729abcc0a57eee5db8a7a449c00f2a5", "generation_seconds_batch": 16.6739911820041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2500059604644775, "target_z": 2.2343749831771733, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.7608399391174316, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b0009b1d4e9b816b66663fbd9eeca50296a511fee445ab444d6aed07f7cb70b.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0b0a5ed8d226a26a3c509233f2b9d43b1c1237b1c03faa8947d7d31e4a03a1bd", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "661dbd131a197173c3d5055b8ba5107b5ca14400ec136dd89b25413d371657e0", "generation_seconds_batch": 16.6901885290863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.41180419921875, "genuineness": 4.3881025314331055, "identity": 0.5365066528320312, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b0a5ed8d226a26a3c509233f2b9d43b1c1237b1c03faa8947d7d31e4a03a1bd.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0b5ea810169331653fd7d327babf449042cae4e11a50eb6e97819a787ff54437", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fd2bf76712d847263f61b95ae57344787845780a47652be12ce58223a0f8843", "generation_seconds_batch": 23.44470277102664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 6.427689552307129, "genuineness": 4.532451152801514, "identity": 0.47823217511177063, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b5ea810169331653fd7d327babf449042cae4e11a50eb6e97819a787ff54437.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "0b7de4c5288cad2f0b528df2a6cbf48c5c4ba9044bc5df0dcdabcf456e0b46d5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29fcb4c71cc77fdc7c1d6007cac43357d93ddc06eebdb4890c45b81fb6c921e0", "generation_seconds_batch": 22.92463071702514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.229123115539551, "genuineness": 4.306513786315918, "identity": 0.5308403968811035, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b7de4c5288cad2f0b528df2a6cbf48c5c4ba9044bc5df0dcdabcf456e0b46d5.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "0b810fd3385800272ab89bd17e91a1ce64002cf3c15e775ed6b348a9b1b29cbf", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fafd11c0eda215f3ab94d492db9b1f90c3ccf2bc9396cd26facbe0b6e5b6fb0", "generation_seconds_batch": 17.125282229972072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.284721851348877, "genuineness": 3.9676074981689453, "identity": 0.36286115646362305, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b810fd3385800272ab89bd17e91a1ce64002cf3c15e775ed6b348a9b1b29cbf.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "0b86fbcdbe062ae840d5d721c1c2cee8e023f8a1adc49e48e4ea6c3265dc8812", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b6f35c6d1b6e06a2d1b5ae2122b92882f3804f606abc96929178b763b55455c", "generation_seconds_batch": 20.91654931695666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.006728172302246, "genuineness": 4.582659721374512, "identity": 0.5807204246520996, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b86fbcdbe062ae840d5d721c1c2cee8e023f8a1adc49e48e4ea6c3265dc8812.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "0b8e3ff8ff5a65d24e783ec00603d7e94fe645fd882572253f3270bebf52bfc2", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79d2c2f5548df7c858b738c4571b12f5deb3ab28bae3204a6b21243950952cb7", "generation_seconds_batch": 16.613817290985025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.165581703186035, "genuineness": 4.482243061065674, "identity": 0.5032440423965454, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b8e3ff8ff5a65d24e783ec00603d7e94fe645fd882572253f3270bebf52bfc2.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "0b91ec945b859bae3268439515e675d7f4a4a9767979a1d068531ac8766a5a46", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0227cb0bc56fc7ab6c3f2125b9fd7107f345cf8c42d8396749d50ec4dd34a9a2", "generation_seconds_batch": 15.695993277942762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.300607204437256, "genuineness": 4.457139015197754, "identity": 0.6920504570007324, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b91ec945b859bae3268439515e675d7f4a4a9767979a1d068531ac8766a5a46.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "0b94470d10af25aa4249a96516e751950306827777c744f35a1103e91c45a21f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75ae6d7c760474d35f33101bc5a78d8f6202c4f06ef8f19f807c27cb70a65498", "generation_seconds_batch": 22.997103448957205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9178904294967651, "target_z": 1.1406250290576094, "ce": 6.157639026641846, "genuineness": 3.8169827461242676, "identity": 0.37564072012901306, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0b94470d10af25aa4249a96516e751950306827777c744f35a1103e91c45a21f.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "0bc5c593021a9ed3d8942233cd19b7c4471e34b1a3dda22846610a86f25e33e7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "036b136864ff50d57d7fcdf3f42973c713cd40d4ffffdccff2ed80bd13eb7013", "generation_seconds_batch": 23.143088513985276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.348263263702393, "genuineness": 4.13078498840332, "identity": 0.6592683792114258, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0bc5c593021a9ed3d8942233cd19b7c4471e34b1a3dda22846610a86f25e33e7.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "0c70faee74b29feff203ff25e2f47beb48dc82709edf1bc322739901670d5d87", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0867f904a698e4d1e489cc1e794b60c78f20ced0416644f493f5f0aa8f3b9dbd", "generation_seconds_batch": 16.78895953495521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5744330883026123, "target_z": 1.679687509940761, "ce": 6.308549880981445, "genuineness": 3.8420867919921875, "identity": 0.26584941148757935, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0c70faee74b29feff203ff25e2f47beb48dc82709edf1bc322739901670d5d87.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "0ca81c81943933d9aa2bbeceeef4978c3be9c77ccc0f7ebabcbb7c56f178b42b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29775d0596a9fb82e4f4f2d3b33c3e4b16c7c075d585cdcc9a1cd5689abf45f5", "generation_seconds_batch": 26.450942093040794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.054384231567383, "genuineness": 3.7667744159698486, "identity": 0.5955508351325989, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ca81c81943933d9aa2bbeceeef4978c3be9c77ccc0f7ebabcbb7c56f178b42b.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ba583f6ea8526af255bf3d39afbdc89e2249b6b7613a0181efc917d0ab10324", "generation_seconds_batch": 16.8535391719779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.149696350097656, "genuineness": 4.381826400756836, "identity": 0.7642890214920044, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0cc33276b18f11868bb7ef544c3b5be61da0c1b6cfb50c61f58d0561898fb29c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc62b89a4f6687b77aeaa661c886f93278eb97b46c158eb51a61097dc26e41ec", "generation_seconds_batch": 16.627347600995563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.316492557525635, "genuineness": 4.074300289154053, "identity": 0.5921361446380615, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0cc33276b18f11868bb7ef544c3b5be61da0c1b6cfb50c61f58d0561898fb29c.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "0cd58cff3ef7b4b8da0fc44090f06e8ccdd915c40b890cc07137a92c60d36c6d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f001013035084ce40e76818aa57d186d5c7170bf1ddbcfe391bd125d5505ce1", "generation_seconds_batch": 21.53497124405112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.070269584655762, "genuineness": 4.293961524963379, "identity": 0.5721262693405151, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0cd58cff3ef7b4b8da0fc44090f06e8ccdd915c40b890cc07137a92c60d36c6d.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "0cdc1d0fbbd198ec9044235fe8e964d7d9927ec75b95ff7661f9ac740fb79b8f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9a8978a24ef2212768097c0560b626545036a8e3459de9d9547642c02c9ad9a", "generation_seconds_batch": 17.870354893035255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.898860216140747, "target_z": 1.125000036704349, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.5990855097770691, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0cdc1d0fbbd198ec9044235fe8e964d7d9927ec75b95ff7661f9ac740fb79b8f.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "0d8349f08d051e7f11e2168dfa6ce0349ebf389e85d79b1a193eebeb1cda32c4", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30a117ea360a10632e9318f60d96f1463ec3ad0548710f6d19c23a6d71ffd808", "generation_seconds_batch": 16.386504685971886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.7290402054786682, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0d8349f08d051e7f11e2168dfa6ce0349ebf389e85d79b1a193eebeb1cda32c4.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "0d869d2182dfd341759f73bd5817f6cc0a858d0861eeb73fb0a2fdb0faad8595", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "70404c30a884db46e73bb53f6ecba5894c0d4b35b6053f6303df9e2fe4e88407", "generation_seconds_batch": 15.481736693996936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.7593821883201599, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0d869d2182dfd341759f73bd5817f6cc0a858d0861eeb73fb0a2fdb0faad8595.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "0de5d1893084869480356d6291ba633e7d2aebd424697684e4e397b9cbf6af16", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63f53f47ec5f96f4f2ef788d9658f7a512e9a22b917c8b889419cb39a75d5b5b", "generation_seconds_batch": 17.41624158900231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.157639026641846, "genuineness": 3.9299514293670654, "identity": 0.7537817358970642, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0de5d1893084869480356d6291ba633e7d2aebd424697684e4e397b9cbf6af16.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0df9d9fb4b56d75d9e5d5c4c17e44dc0158756267be75b009b9a7961811aef0e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd2141d3869d9bdd0ef3ddd4eef44dca15d4b1d7de4f4319b56b7d7e64f79984", "generation_seconds_batch": 21.637516026035883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.173524379730225, "genuineness": 3.8797430992126465, "identity": 0.3090677261352539, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0df9d9fb4b56d75d9e5d5c4c17e44dc0158756267be75b009b9a7961811aef0e.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "0e53fb0581157d88c6780f27f13701bde27e1e2e0350d34b195ff003431912cf", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4c8d57099d4c0b55e53598bcf4a9d8081e3b6ee0fe6362b49875fefb67fbab8", "generation_seconds_batch": 21.500453840009868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.205295085906982, "genuineness": 4.118232727050781, "identity": 0.36935877799987793, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0e53fb0581157d88c6780f27f13701bde27e1e2e0350d34b195ff003431912cf.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "0e8968882263eb2272f972edb08110f6f4336c8fa69941dcf84bf7da4d802217", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54fa7c305969daca701dd0e5fe8a09ffee1d190d5a1d82010e20f5f0ff86ace3", "generation_seconds_batch": 23.08624123397749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9178904294967651, "target_z": 1.1406250290576094, "ce": 6.260893821716309, "genuineness": 4.21237325668335, "identity": 0.5243438482284546, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0e8968882263eb2272f972edb08110f6f4336c8fa69941dcf84bf7da4d802217.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "0e92f21e439e6af17f23782c2d97d88490f5143ce3ad80f2feb7d1cd1548ebcf", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d59dedcda31dbd8848fb74ec4a4d132274a35553a3abdbbb5a51a30804f4ff98", "generation_seconds_batch": 22.39065151894465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.443574905395508, "genuineness": 4.193545341491699, "identity": 0.40626296401023865, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0e92f21e439e6af17f23782c2d97d88490f5143ce3ad80f2feb7d1cd1548ebcf.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "0ea42a31279444eab9b919f9df418e5bbf888096244232728da713b2d782ae35", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61fe2bc6270d2c16e3bcd755c820323638cefc87ee0b5b14399e8a017f954e86", "generation_seconds_batch": 17.870354893035255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.2767791748046875, "genuineness": 4.256305694580078, "identity": 0.4092433452606201, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ea42a31279444eab9b919f9df418e5bbf888096244232728da713b2d782ae35.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "0ea5f4dcf7c7d4727bbcaa171704a74a61cd04cb8d20e4b3c4193a873126c368", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f63ad13e4d969be42646e1fa7d983fad4559e0a7583f7b61d61165cd566dc33b", "generation_seconds_batch": 18.712021928979084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.8542017936706543, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ea5f4dcf7c7d4727bbcaa171704a74a61cd04cb8d20e4b3c4193a873126c368.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "0ef6aaab6832bc8c32070f5be5b62e0296452fb644cac8164a1f3f14e1b2633c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85fd2727cd5bb20e0b0d26fbb4973d0046fd681970cc56a4fc3eddb988af22d0", "generation_seconds_batch": 17.254525818047114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.022613525390625, "genuineness": 4.306513786315918, "identity": 0.7173230051994324, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0ef6aaab6832bc8c32070f5be5b62e0296452fb644cac8164a1f3f14e1b2633c.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "0fa32af3c6192e03192b1a143d03f2efcbe7bedf1ef4da6128f2312938856ff1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9e76d9ae5b9461129149ccea12d0a969484243a4e1f4271d330a8e438d9885f", "generation_seconds_batch": 26.42855558008887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.0146708488464355, "genuineness": 4.017816066741943, "identity": 0.5551449060440063, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0fa32af3c6192e03192b1a143d03f2efcbe7bedf1ef4da6128f2312938856ff1.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "0fb03ab5340323c8987510bba762c2f4ac650f689d55117a0af94b520b088331", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8165fc3f0839326d1d7761e9050ed0ed7b29591fdd418799e920d4986aae6b08", "generation_seconds_batch": 23.46767769707367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 5.935244083404541, "genuineness": 4.432034492492676, "identity": 0.5958545207977295, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0fb03ab5340323c8987510bba762c2f4ac650f689d55117a0af94b520b088331.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "0fb826b81f06fc9ae2adaa5c9849bcc49b15f65e665a9c655cba2c6b7cd298c0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8989060e98db7b0141cf3d207c344893e945b3a3cd32681fbce39d798d6fa681", "generation_seconds_batch": 25.688200602075085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.348263263702393, "genuineness": 4.18099308013916, "identity": 0.6683349609375, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0fb826b81f06fc9ae2adaa5c9849bcc49b15f65e665a9c655cba2c6b7cd298c0.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "0fd3fc9fb0b6b7638f5cb3bdae0a60d32b37858b3e085b02a003945a3c1835bf", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ba5014e5e5730796b0358a0f62836a855739fa523261a182075b832a1c14f5ac", "generation_seconds_batch": 16.60845820605755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.348263263702393, "genuineness": 4.331617832183838, "identity": 0.632086455821991, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/0fd3fc9fb0b6b7638f5cb3bdae0a60d32b37858b3e085b02a003945a3c1835bf.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "102e58bfcd79b8bfd8ce36fb16ca7ab6d01356a35409f5018b863dcd6de75721", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da72f6b411e5fe80dae5eddbddd43844ea3867b55a87f667ce9386cbcd7da680", "generation_seconds_batch": 23.13862421794329, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 5.959072113037109, "genuineness": 4.268857479095459, "identity": 0.6995077133178711, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/102e58bfcd79b8bfd8ce36fb16ca7ab6d01356a35409f5018b863dcd6de75721.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "10401a02fba3f939f7b9d86d7d967e71f961ffa6a284b217dcc70ef863c09e5d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abad5611716d017dcafb5e06f1434b545b91790844d1ba9ed90c680360fb78ed", "generation_seconds_batch": 24.52314870490227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 5.847874641418457, "genuineness": 4.444586753845215, "identity": 0.804313063621521, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/10401a02fba3f939f7b9d86d7d967e71f961ffa6a284b217dcc70ef863c09e5d.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "105610bdee1a763e119bfcbff6619ca64385f5008ee94db8151f9b86aed91f63", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "199dd4ab0e9b00f813b5704f74e3241c96142632720dc8efcc494a301de9dc14", "generation_seconds_batch": 22.97154399496503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.348263263702393, "genuineness": 4.2249250411987305, "identity": 0.41800349950790405, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/105610bdee1a763e119bfcbff6619ca64385f5008ee94db8151f9b86aed91f63.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "112fa06ef94f8428322ee055a95b9a4a93dc1676159a9799ac66643bac1480da", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8752632466c765fe85d8d64e0caa20fc90071e3a0627528b401e5c20ba95f36", "generation_seconds_batch": 24.948357441928238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.7998946905136108, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/112fa06ef94f8428322ee055a95b9a4a93dc1676159a9799ac66643bac1480da.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "117c804a5e903ffc6638120dffaf0bc78a82fb6fff4f014748991e630960d74f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ff2e21f71874fd180a9093b23dfe950ea43f2bd32a2218363476e80906859045", "generation_seconds_batch": 19.25936669902876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.070269584655762, "genuineness": 4.218649387359619, "identity": 0.6147832274436951, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/117c804a5e903ffc6638120dffaf0bc78a82fb6fff4f014748991e630960d74f.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "119bbc6745b7d12c0fa4ffa8854ed639897449aa0108faf95f70976e1bd9d9ab", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0f265e18394b46c2e28a75bf1bc851d976fce8c52db47a50c1aafb86e03bfd36", "generation_seconds_batch": 24.668218436068855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.252951145172119, "genuineness": 4.162164688110352, "identity": 0.5631998777389526, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/119bbc6745b7d12c0fa4ffa8854ed639897449aa0108faf95f70976e1bd9d9ab.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "11a70bc543068096202574e18e1ac432d591ca54c82c3b957e97f4da0c787c8d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1db1e5e044b894842e2222dc2aa18d89f86a0091532a23f0803baf6b39f740d5", "generation_seconds_batch": 17.91126184805762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.136737823486328, "target_z": 1.3203123921809754, "ce": 6.41180419921875, "genuineness": 4.482243061065674, "identity": 0.30949679017066956, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/11a70bc543068096202574e18e1ac432d591ca54c82c3b957e97f4da0c787c8d.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "11a97fd953e53ee9c243b787696306a8ed147919cc73367f691f63195d709a8b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "346be2ecf7bb9567a8292be95152e8adb6f4d850103d01a47830193c7deacfcb", "generation_seconds_batch": 18.594492787029594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 5.486483573913574, "genuineness": 4.444586753845215, "identity": 0.665770411491394, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/11a97fd953e53ee9c243b787696306a8ed147919cc73367f691f63195d709a8b.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "12974a085e3c8e4b7d72baae77742ddec5c5ce19ce9de610ac3b1680e96630bd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdfe5d0685b7bff43573ec7b473b710203376b153f57beaf5813a4e34e094bcb", "generation_seconds_batch": 24.195480598020367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.125868320465088, "genuineness": 3.923675537109375, "identity": 0.6882216334342957, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/12974a085e3c8e4b7d72baae77742ddec5c5ce19ce9de610ac3b1680e96630bd.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14f1962b033d97568de44a43121a6da5c1fc264d1d2086d8df6b0de551f550d8", "generation_seconds_batch": 17.06627894495614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.141753673553467, "genuineness": 4.080576419830322, "identity": 0.7197281122207642, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "13964f715f8f94fcf575a859324e70028599432732461e67ea19ba4b4e020aa2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "399af75789fb06bf973cfafe1101b13b9baefe44e1602155d1267adb9a33e310", "generation_seconds_batch": 17.200013309018686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.268836498260498, "genuineness": 4.118232727050781, "identity": 0.6315615773200989, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/13964f715f8f94fcf575a859324e70028599432732461e67ea19ba4b4e020aa2.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "13b33a97f970e31a3906e7f054eb2a96b2ec1d5bdc5c1c1a20bd9c72c8798226", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27e94dec6e52ca63e46c8ab451829a13f9f87501ab874e0ccbb65e620a55cef5", "generation_seconds_batch": 16.805235796025954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.062326908111572, "genuineness": 4.400654315948486, "identity": 0.5969969034194946, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/13b33a97f970e31a3906e7f054eb2a96b2ec1d5bdc5c1c1a20bd9c72c8798226.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "13b9305328e958aa66f6402ffee280ec0d8396333ac92a635325ff6a801e6954", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15c192ebff1736c08787934c3e218b2c5b627b9de87bf68d3c5f1abcb0674e67", "generation_seconds_batch": 22.704402297968045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.7032938003540039, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/13b9305328e958aa66f6402ffee280ec0d8396333ac92a635325ff6a801e6954.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "145a100ff34d1b72056d9f71d2e719c814d600a8fbde01fed3e2f9ecbb93516b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6f6e087fe8e6b0026a439861e6e58382644c216e2be8a9de375e69e4d405df0", "generation_seconds_batch": 22.80701476393733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.356205940246582, "genuineness": 4.231201171875, "identity": 0.3899611830711365, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/145a100ff34d1b72056d9f71d2e719c814d600a8fbde01fed3e2f9ecbb93516b.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "14b0e56a4f4916315bc29ea5eb48e64d7f59368172dd6833cb84c3a442422428", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bef5f5732646d350baf3f1761660300362ca90f36acc2487efc619fbb13961bb", "generation_seconds_batch": 19.062100211973302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 5.959072113037109, "genuineness": 3.9425034523010254, "identity": 0.5661802291870117, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/14b0e56a4f4916315bc29ea5eb48e64d7f59368172dd6833cb84c3a442422428.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "14fb64f500cfd8e1bc9080f9a99baf75af83b54328030c027bea80a9470aa91f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c73c1f3d7ae862bb1c438e47daa6ee96a141d079ff215017d57bd08b171b407a", "generation_seconds_batch": 23.25263145903591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.141753673553467, "genuineness": 3.948779582977295, "identity": 0.39089855551719666, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/14fb64f500cfd8e1bc9080f9a99baf75af83b54328030c027bea80a9470aa91f.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "151ea8493df1737901b739ecaeabc426092917b37168d244e3f4c36f917845c4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e55721fb961478577d2a440c6677b28fc0b9e65aada407ca5383acfde6de456f", "generation_seconds_batch": 17.41624158900231, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.7392585873603821, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/151ea8493df1737901b739ecaeabc426092917b37168d244e3f4c36f917845c4.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "153a11c8b2371cb30dfc988a6c5c2983bdb48bea3fbc7cbe1328190c19980f2f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7aa2c3f73ba5e64064864689a6677dfdba9f63e40a52a704bf84477251d222d2", "generation_seconds_batch": 24.290733593981713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 5.895530700683594, "genuineness": 3.8671910762786865, "identity": 0.5529400706291199, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/153a11c8b2371cb30dfc988a6c5c2983bdb48bea3fbc7cbe1328190c19980f2f.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71288b96096edb80cbd0b990cbcf3cd481ad9b7979e485d2f0bce494019d2770", "generation_seconds_batch": 16.96080718608573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.467403411865234, "genuineness": 4.030367851257324, "identity": 0.3801848888397217, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "157a466853efef135ae02df90d9812d2a38d997d3b50e3abd2db952232ba10f3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c16a88ff0a79f9b7508e01f754914344a838ea55fbb59781297506a1768266d", "generation_seconds_batch": 15.109691446996294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.443574905395508, "genuineness": 4.2751336097717285, "identity": 0.4323456883430481, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/157a466853efef135ae02df90d9812d2a38d997d3b50e3abd2db952232ba10f3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "168835d593f74819b375226dc661ac4a38654a84a02360de6237c4d05a01ccd3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fad2cc6ee6159137fc982fee25177ad814470640f6c9eedadd5a7cfcbbea54f", "generation_seconds_batch": 17.481961090001278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0415868759155273, "target_z": 1.2421875282929355, "ce": 6.046441555023193, "genuineness": 4.457139015197754, "identity": 0.7000476121902466, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/168835d593f74819b375226dc661ac4a38654a84a02360de6237c4d05a01ccd3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "16a047db1d58da9b68a34a03aeb9dd1eabbe3099a370fc0fa5f45cba3854b87f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c29926f1899a639b871704f516e1abab2a78356668c164aa82a07725267f3c02", "generation_seconds_batch": 22.11369329004083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 5.919358730316162, "genuineness": 4.375550270080566, "identity": 0.6355577707290649, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/16a047db1d58da9b68a34a03aeb9dd1eabbe3099a370fc0fa5f45cba3854b87f.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "17189cbd00d622fc22659e6f6aa7a0c73c891ab7d4b218b3f49b3ab235a1a26e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d00bec9637f22bd57f1b934c6aeee27e881c9e392e4b48f283b993d4e9ee33aa", "generation_seconds_batch": 23.791357461945154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.660982370376587, "target_z": 0.9296874854711953, "ce": 6.157639026641846, "genuineness": 4.256305694580078, "identity": 0.6144677400588989, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/17189cbd00d622fc22659e6f6aa7a0c73c891ab7d4b218b3f49b3ab235a1a26e.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "17598a22114083b2b5bc03bb1eacb845fc7231f69c415fd82a7fedb1fd99f749", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48e6a21c9628d5dc2fb044996784fdeaf81df96f01d2d0390611dd5ff187dd3f", "generation_seconds_batch": 23.782389735104516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5753463506698608, "target_z": 0.8593749709423906, "ce": 6.300607204437256, "genuineness": 4.400654315948486, "identity": 0.4637901186943054, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/17598a22114083b2b5bc03bb1eacb845fc7231f69c415fd82a7fedb1fd99f749.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "1785d09db88717f6cbea39c9b992d77fda02c36e82c6c67a480b3de30a1e2f31", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5779645b61bc8164f520e32665e23be9384613e7e0b815306c7ef24a2a90f37a", "generation_seconds_batch": 16.63327646604739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1747984886169434, "target_z": 1.3515625726440237, "ce": 6.427689552307129, "genuineness": 4.293961524963379, "identity": 0.6819942593574524, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1785d09db88717f6cbea39c9b992d77fda02c36e82c6c67a480b3de30a1e2f31.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "17b0046d5e59dbaed187ec12c5c35ca1e56b9193952605a2039b3d0120532ca0", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebfc15cc8ae94939506e291e5fa654f0f2c44b4f5562e88a5c74d15699d0c9d2", "generation_seconds_batch": 16.68164375203196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 5.9431867599487305, "genuineness": 4.344170093536377, "identity": 0.4856928884983063, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/17b0046d5e59dbaed187ec12c5c35ca1e56b9193952605a2039b3d0120532ca0.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "17ba8f372247282e440a6aaa44fad6bb719c33da92cf446b003644af308b42fc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4d3a4c85f5d13770f21098fbee51621235c5096da06933295a1b07424c7eed6", "generation_seconds_batch": 26.883209926076233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.046441555023193, "genuineness": 4.306513786315918, "identity": 0.7569718360900879, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/17ba8f372247282e440a6aaa44fad6bb719c33da92cf446b003644af308b42fc.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "17ec39c46eeec5785b38a9a3a20d9c10488c6518ff805625971063e5c594edcc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f866795c13455b6ac80e95b524fef640637ee6b85973329d2b36e41b76a57815", "generation_seconds_batch": 17.098034358932637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.1894097328186035, "genuineness": 4.2751336097717285, "identity": 0.7482904195785522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/17ec39c46eeec5785b38a9a3a20d9c10488c6518ff805625971063e5c594edcc.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "180fb980d305290c9344727fee67ce3d8dea9432e4aebdb0438f1f25f48c4f81", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a86ede863a36317afce458d2525c047477d023d8ea6fff1d2e822adf4ada484", "generation_seconds_batch": 23.31823790294584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.41180419921875, "genuineness": 4.231201171875, "identity": 0.48845624923706055, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/180fb980d305290c9344727fee67ce3d8dea9432e4aebdb0438f1f25f48c4f81.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "18b9741d01270109a711af41de2488990b6eef2929e90b8e0fd8fec841f03dce", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "653bce06c0c492f51aa94e9535721bbf8723d5d9374378e2ec26f18d501ca697", "generation_seconds_batch": 16.50531110400334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8417694568634033, "target_z": 1.0781249617663033, "ce": 6.252951145172119, "genuineness": 4.457139015197754, "identity": 0.2210068255662918, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/18b9741d01270109a711af41de2488990b6eef2929e90b8e0fd8fec841f03dce.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "18c792d870f8e3102b4282ec36463a2e0ba0366dba46217ae9714af4db941ad2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b537c3312d68b2f3782f0f7f3089d73f2ded28c4396e0b74ad73ca624ffee2d", "generation_seconds_batch": 23.27275126997847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7751637697219849, "target_z": 1.0234375374690228, "ce": 5.982900142669678, "genuineness": 3.9676074981689453, "identity": 0.6163091659545898, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/18c792d870f8e3102b4282ec36463a2e0ba0366dba46217ae9714af4db941ad2.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "18ca1fb923c577f289b5e4f3ffa6bf5dcb0cf015954c3824b7e10f53ed62182e", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81fce1e8aae50455eac8110ce7e887c654b45080ff3c76fe8e24ed546e51f46e", "generation_seconds_batch": 17.472881059045903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.269949436187744, "target_z": 1.4296874365320635, "ce": 6.0305562019348145, "genuineness": 4.381826400756836, "identity": 0.5986768007278442, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/18ca1fb923c577f289b5e4f3ffa6bf5dcb0cf015954c3824b7e10f53ed62182e.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "18efc31034c9627558b5a5c0b7a127872988af12938427b192e54b8c1b640495", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63aeee4be3c122f2f57f6c329de9ba9f2f56acd4e0c3736c299a4bb90381b103", "generation_seconds_batch": 16.47782673605252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.38797664642334, "genuineness": 3.8546390533447266, "identity": 0.24113214015960693, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/18efc31034c9627558b5a5c0b7a127872988af12938427b192e54b8c1b640495.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "1930d820e7a22c586888d9d93432b6ec76c56f570f4406c3d3ab288dadb458b8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2a900f9fa7529c729d97c3fd0e8ecf4d2e088961262d0ff72d88c3e3653e359", "generation_seconds_batch": 20.11246775498148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 5.967014789581299, "genuineness": 3.9299514293670654, "identity": 0.3406468331813812, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1930d820e7a22c586888d9d93432b6ec76c56f570f4406c3d3ab288dadb458b8.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "193fb5c59a755740dac74c5cc66fa038ab2b66b52a50382e83d5bea2d593d09b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7014426d95e710081f527b16b77910953940350ec6c0192f3b9661e0bb024e1", "generation_seconds_batch": 23.678506216034293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269036293029785, "target_z": 2.250000073408698, "ce": 6.0146708488464355, "genuineness": 3.8797430992126465, "identity": 0.6644347906112671, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/193fb5c59a755740dac74c5cc66fa038ab2b66b52a50382e83d5bea2d593d09b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "1954a0e783f067957c916d1ac5000f66af008efc412039fc6bdeed21339f683c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec0774e9213566d15cc7fc256c7b1a6d51e86ad8f75f6936456d6e383eabb975", "generation_seconds_batch": 23.870291389059275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.252951145172119, "genuineness": 4.030367851257324, "identity": 0.6518812775611877, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1954a0e783f067957c916d1ac5000f66af008efc412039fc6bdeed21339f683c.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "19fff28054cc2713f09f2f14851b440186f39d6cc38cb26c16be5104a6ec2405", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4e7a2a1814005ca5d90561ec2ba267765098fa073b5b8486458f0e5b3c7cef85", "generation_seconds_batch": 16.534983063000254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.340320587158203, "genuineness": 4.024092197418213, "identity": 0.4467311203479767, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/19fff28054cc2713f09f2f14851b440186f39d6cc38cb26c16be5104a6ec2405.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "1a3ac3d8953ff8a6b39ed4039dbb500fe29951002baf18c9c39277af45fc975c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d291edd350e2e86b9c8a3f0d505c4946514283a5f39b96d63729b70ab88736a", "generation_seconds_batch": 16.60845820605755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.292664527893066, "genuineness": 4.042920112609863, "identity": 0.5045755505561829, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1a3ac3d8953ff8a6b39ed4039dbb500fe29951002baf18c9c39277af45fc975c.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "1a46df677e12737dbb6a9095ee04b6df370af81486dee1bb12be504ddcd39741", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6309ecdaa373a3b556664b70cf0805ac80cb512a2bd547b827d8929c3b7a2940", "generation_seconds_batch": 24.047121389070526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7180730104446411, "target_z": 0.9765624625309772, "ce": 5.442798614501953, "genuineness": 4.557555198669434, "identity": 0.6841151714324951, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1a46df677e12737dbb6a9095ee04b6df370af81486dee1bb12be504ddcd39741.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "1a79004deaff22d08c9d39d9e727f11f7011cc29a3f0d364e54241644eff35f3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8df4345835c5322ab57701997ccaeae0171fdbff9e17b0ccb6813a2c00337bfa", "generation_seconds_batch": 20.795082685071975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.38797664642334, "genuineness": 4.105680465698242, "identity": 0.6103358864784241, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1a79004deaff22d08c9d39d9e727f11f7011cc29a3f0d364e54241644eff35f3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1a9656e114f6eb75c8bfabaadfb1b81217c34a9700eb1feb5ef98d04cb7bb279", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d188fec7033f6703fb1e7dd5aaea027072193a5974fed2317508a6ac7f972a68", "generation_seconds_batch": 20.16824153799098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.6620680689811707, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1a9656e114f6eb75c8bfabaadfb1b81217c34a9700eb1feb5ef98d04cb7bb279.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "1ab15b8d96bc34ec61606ac01e4582fd0fd821d69c9057760b756d66bfb055c0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8576928a38ed8c49040dfe3db00f436c58606a3366f6b23f5228e1dfdadcf33", "generation_seconds_batch": 22.805018635001034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3451571464538574, "target_z": 2.3125000428217404, "ce": 6.181467056274414, "genuineness": 4.293961524963379, "identity": 0.6921023726463318, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1ab15b8d96bc34ec61606ac01e4582fd0fd821d69c9057760b756d66bfb055c0.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "1acab8b1fb131cf05aba5e3810b7a7d036b3a782d0649595557e5864bac2b575", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a791c9c1b2f10a7f10900ca2c01990c3212f09744fa3e3031c79c9782f37b6e", "generation_seconds_batch": 21.58070328796748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 6.038498878479004, "genuineness": 4.306513786315918, "identity": 0.7744360566139221, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1acab8b1fb131cf05aba5e3810b7a7d036b3a782d0649595557e5864bac2b575.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "1aea2ba3a32ba66a9607b2bdd4bb9825968fd07de6bca3d7b145fc4c3ac40d40", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dca188c10f31f37cddbc20054541ab9c7c2834d39c03e6d1d429f17ed9cf869f", "generation_seconds_batch": 17.042026398004964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.403861999511719, "genuineness": 4.337893962860107, "identity": 0.3552062511444092, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1aea2ba3a32ba66a9607b2bdd4bb9825968fd07de6bca3d7b145fc4c3ac40d40.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1b726cc1e3a087898576d798930b683ba975a2e9eb5fb8a45208083c7f84b377", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32413eb5f541966a086da7a77120b92188c47e82fbd1b6dc42d38b615dde0499", "generation_seconds_batch": 26.620400486979634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.181467056274414, "genuineness": 4.15588903427124, "identity": 0.8331326246261597, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1b726cc1e3a087898576d798930b683ba975a2e9eb5fb8a45208083c7f84b377.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1bb2eaf035f695963aa30baabbd42ed6d983deefd3d04cda575749e1f6ab5a00", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca2a8067b3a36778f02dc66af6232a4833f4dcd21756a279cddf0ff454a8ec14", "generation_seconds_batch": 23.576122549013235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.427689552307129, "genuineness": 4.293961524963379, "identity": 0.614755392074585, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1bb2eaf035f695963aa30baabbd42ed6d983deefd3d04cda575749e1f6ab5a00.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "1c056d2a1c56134b16f4d45072d602cb42d63c38812ebf55ee456fc8173c2c70", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3dfa1e5ea48ae451c7233fa415bf22fd8ce3a956f7b9cc8a998f7733e63f1404", "generation_seconds_batch": 16.25351162010338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1177077293395996, "target_z": 1.3046874977059781, "ce": 6.252951145172119, "genuineness": 4.469690799713135, "identity": 0.33532506227493286, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1c056d2a1c56134b16f4d45072d602cb42d63c38812ebf55ee456fc8173c2c70.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "1c367811a20dca80f437f15c27a9c5bb66950f8830642f336861fe47732fb893", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f53af5e93a465684ff8822820dab270adb5e3260ec8d9c8dbf4bf23a6d4b3d4", "generation_seconds_batch": 17.6602623880608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.5352200865745544, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1c367811a20dca80f437f15c27a9c5bb66950f8830642f336861fe47732fb893.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "1c563a37d512b417e9881dbaa45b2b036a920ed2613dabd864fc12dca55b66a1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cb7c265e82a2c177ccfde33f3837f08ac7930106c07ee28f542ea0fc015e9d50", "generation_seconds_batch": 16.105830463115126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.316492557525635, "genuineness": 3.9299514293670654, "identity": 0.34217971563339233, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1c563a37d512b417e9881dbaa45b2b036a920ed2613dabd864fc12dca55b66a1.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "1c5fc748d21c7c4a2aa2c5a6954c89e3cbd0dee7a7f105652544c83fd062854a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d34fddd79a7e592849fe6633bc14bcf1e555490ba952e9d9c48bd3f11d456fea", "generation_seconds_batch": 17.206857195007615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7457051277160645, "target_z": 1.8203125389983705, "ce": 6.0146708488464355, "genuineness": 4.124508857727051, "identity": 0.6621197462081909, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1c5fc748d21c7c4a2aa2c5a6954c89e3cbd0dee7a7f105652544c83fd062854a.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "1c7f88c494dd5438e37fb2afcb2b6d32281b1a5f53f1cbf53072b2776c061065", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "652865e0d81878178795ff3702e758a8fe57860f872a35d16ee91ff58a3432f6", "generation_seconds_batch": 27.57249904202763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.5017629861831665, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1c7f88c494dd5438e37fb2afcb2b6d32281b1a5f53f1cbf53072b2776c061065.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "1cb84da764946e33dc1ec790e65b5036618027ef50ad1ba32c8b45eed8422785", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8ca93650ce6b8a9309f4fe6e565526b14e9f23af695d32a5717459904923eea7", "generation_seconds_batch": 18.086886983946897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.660982370376587, "target_z": 0.9296874854711953, "ce": 6.403861999511719, "genuineness": 3.961331605911255, "identity": 0.1866898387670517, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1cb84da764946e33dc1ec790e65b5036618027ef50ad1ba32c8b45eed8422785.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "1cb9d1935440d71e2307050ac344a3743a5b1f315f80e9ff23ded7a01201ab92", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "123e0e0b8f246d782d05cd8702fe9985aa5f6a72fe45abad92955e5fbeb04c22", "generation_seconds_batch": 16.993433407042176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 5.998785495758057, "genuineness": 4.300237655639648, "identity": 0.45886069536209106, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1cb9d1935440d71e2307050ac344a3743a5b1f315f80e9ff23ded7a01201ab92.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1ce44f3a9841e629de5186d0db73178da66a9db3bdaea9619489c6632c018c7c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f449a2efa0247f4483fe85bfbefa6442cca371322be64474d5004330a4596654", "generation_seconds_batch": 23.95931884809397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 5.998785495758057, "genuineness": 4.042920112609863, "identity": 0.7743311524391174, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1ce44f3a9841e629de5186d0db73178da66a9db3bdaea9619489c6632c018c7c.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "1d54b61425645c9da66b14876b317810627eebbdbaa38873c26b4f4497a05e2e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ca71475cd19faecbe698b0aac29dc5b5ed82487b24f7a6596d3b4bc9a8cc2d4", "generation_seconds_batch": 15.444292893982492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.497398853302002, "target_z": 2.4374999816478256, "ce": 6.23706579208374, "genuineness": 4.111956596374512, "identity": 0.7330001592636108, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1d54b61425645c9da66b14876b317810627eebbdbaa38873c26b4f4497a05e2e.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48844156b27deea177092b66439cbb353bb143773ed45930506777f76da31168", "generation_seconds_batch": 17.253087316057645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.141753673553467, "genuineness": 4.325342178344727, "identity": 0.6255137920379639, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d83adf1a57e1c5814a56111ddf328def106c7fafaa1f6b7303cd348902e2b55", "generation_seconds_batch": 16.753720266977325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8607996702194214, "target_z": 1.0937499541195639, "ce": 6.348263263702393, "genuineness": 4.607763767242432, "identity": 0.45529693365097046, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "1d85c215551b447474e131049d903cc648711cac4b3dc4dbfec876ef888c3854", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ad97eed439d7bb82c8742717129d0890e8ae542205d4bdecd11c19ee49eeda6", "generation_seconds_batch": 22.704402297968045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.109982967376709, "genuineness": 4.262581825256348, "identity": 0.521939754486084, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1d85c215551b447474e131049d903cc648711cac4b3dc4dbfec876ef888c3854.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "1dddf9418e66cb7eaad6fb6a3c6a5211c96bc11fd0db4e333f03dfcac19e1d7b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4d043603bb575e67d47b6bbc48f8a0fdc40f71a915b703669dedf23440a027b", "generation_seconds_batch": 20.580536823021248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 5.720792293548584, "genuineness": 4.250029563903809, "identity": 0.6547008156776428, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1dddf9418e66cb7eaad6fb6a3c6a5211c96bc11fd0db4e333f03dfcac19e1d7b.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "1decc76f8c5d47f6923bdbc10ac0d6fab38c5b6f8d9c283a9e90363734abd857", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ab0e69ed78c1e98df69cf12b4dbf2cf475b2735911b51879df584f289625528", "generation_seconds_batch": 23.866066505084746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 5.903473377227783, "genuineness": 4.13078498840332, "identity": 0.7601054906845093, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1decc76f8c5d47f6923bdbc10ac0d6fab38c5b6f8d9c283a9e90363734abd857.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1df59d9e349d897897481d3c8c221caa93127670a47bb1fa31b7e69caa7e81f5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "880171ea3257aa2e33013c34dea142f7c96734e531c1c8242e9108c76c4b4be5", "generation_seconds_batch": 23.07975836796686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.726674795150757, "target_z": 1.8046874487668465, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.6409133076667786, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1df59d9e349d897897481d3c8c221caa93127670a47bb1fa31b7e69caa7e81f5.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "1e55ce4cbef3150d9ae43ac1cf140e6cbc7b97d7d49745ac699d68d124d37430", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c24cbef6ec6f919ce01a071c5ba0559108c217e597d3812886095e1b10a662db", "generation_seconds_batch": 17.13688526302576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 5.903473377227783, "genuineness": 4.118232727050781, "identity": 0.6310344338417053, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1e55ce4cbef3150d9ae43ac1cf140e6cbc7b97d7d49745ac699d68d124d37430.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "1e6c770f1f0e006edb743f9319f55258f47aa77f3d776ca3732f0881f6054289", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb331e21718c22e75a2c588b8774f3dfe16c6e0062e10331a566755d2a9af31b", "generation_seconds_batch": 24.50352984399069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.316492557525635, "genuineness": 4.3504462242126465, "identity": 0.38024142384529114, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1e6c770f1f0e006edb743f9319f55258f47aa77f3d776ca3732f0881f6054289.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Escribí el número en el sobre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f30589c313ae88c6bb6acbf64326fcbd4d4c100427de7fbd2cf0703e5827496d", "generation_seconds_batch": 17.253087316057645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.125868320465088, "genuineness": 4.1998209953308105, "identity": 0.6896552443504333, "hypothesis": "Escribí el número en el sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "1ebd9af1a805310f99d96cc0f5de37fa11e2b7a9414202197d59b14da8d62f5c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d15d2fd326a73851f11ccc69f2cf84955d7c8103f0875cbd8556956a3e941549", "generation_seconds_batch": 22.11369329004083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.165581703186035, "genuineness": 4.319066047668457, "identity": 0.6397032737731934, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1ebd9af1a805310f99d96cc0f5de37fa11e2b7a9414202197d59b14da8d62f5c.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "1f2e895c8dc1452ffb3452c73e06502d306f6083a708462e3320609a6222c979", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41856b237b945f3b56ea882d945b2807c5a77c786bc212fab34b6381258f06e4", "generation_seconds_batch": 25.97485847293865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 6.1020402908325195, "genuineness": 4.18726921081543, "identity": 0.6008999943733215, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1f2e895c8dc1452ffb3452c73e06502d306f6083a708462e3320609a6222c979.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "1f506895adeca9ea8a804280d6a4db40ab5bb490de505f8950de8e76bb0eb0e1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9ded6222a333f8f20ff4df18d0ef36e189b14c23e0cd774d75d22d527ba56d5", "generation_seconds_batch": 17.463447876041755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.157639026641846, "genuineness": 4.494794845581055, "identity": 0.5299341678619385, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1f506895adeca9ea8a804280d6a4db40ab5bb490de505f8950de8e76bb0eb0e1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "1faef11675a63ab0de8f913bfd98933515549bd3b18dc60ef497b05fc728effd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48271f854b9cf41a9b0c69fcd143a185daa9955c96c3a353867fd8d3413c46b8", "generation_seconds_batch": 22.580291702994145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.165581703186035, "genuineness": 4.18099308013916, "identity": 0.6236721277236938, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/1faef11675a63ab0de8f913bfd98933515549bd3b18dc60ef497b05fc728effd.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "2000db92a8e7599e36d9e290abd294786540236b771e1ee8689f97ee18dec122", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "489a7b998e8a6785efcc7db0f773b371e9439f9849be05f0521f1399ea4462b9", "generation_seconds_batch": 16.565797150018625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269036293029785, "target_z": 2.250000073408698, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.6084798574447632, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2000db92a8e7599e36d9e290abd294786540236b771e1ee8689f97ee18dec122.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "203c76fd908a090d0e1c609582d62a78b4df50a09b5b86811aed5dadaf30b772", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1afd2ba032bf958ed8a2ec9e34237c232546587a54a14ba58581415cafb3f12b", "generation_seconds_batch": 17.790041718049906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.205295085906982, "genuineness": 4.218649387359619, "identity": 0.6879846453666687, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/203c76fd908a090d0e1c609582d62a78b4df50a09b5b86811aed5dadaf30b772.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "20751189478a531edcada4824b414ec7c2045da9981f238be77fde5d4b6a8d92", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90b00133398e8d2a8d06d2d4ddbfa05fe8fcb21ef7411281fb8dfc24a4f64b7d", "generation_seconds_batch": 25.800869549042545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.0016390196979045868, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/20751189478a531edcada4824b414ec7c2045da9981f238be77fde5d4b6a8d92.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "2076e2a18ee12c60b12e8a158c2960b8659f484ad3bc560996a7dd6268c997de", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "393dd3af0cb7f094f7915f3e553948915a16fef90018ae834d689df6caf24172", "generation_seconds_batch": 15.541045977035537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.573519706726074, "target_z": 2.499999951060868, "ce": 6.308549880981445, "genuineness": 4.218649387359619, "identity": 0.5923206210136414, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2076e2a18ee12c60b12e8a158c2960b8659f484ad3bc560996a7dd6268c997de.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "20a75cdbd493b2424c2e4430c554009409c02922ae7963180f9b21d646838c17", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "231e9023061d36a413389280f87e11deb29e16bd99037be5ce5714b87a42775d", "generation_seconds_batch": 21.97664772195276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.308549880981445, "genuineness": 3.923675537109375, "identity": 0.5706285834312439, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/20a75cdbd493b2424c2e4430c554009409c02922ae7963180f9b21d646838c17.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "20bc53e6fa0a61c22f9344f37005aede6f08b5772e9b8cc4912dca60d5ea502c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f37900be67c7e257191551f2e07f130033914473c2a3826e62142e3773420b37", "generation_seconds_batch": 27.376497276010923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.197352409362793, "genuineness": 3.9299514293670654, "identity": 0.5792732238769531, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/20bc53e6fa0a61c22f9344f37005aede6f08b5772e9b8cc4912dca60d5ea502c.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "21093ff0e36521c462835e1cf2adac27c092e2b0fef162eb3f95295d511e8b52", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c2e265066de90bbb3c440db87bb3a6b3f57d72b27559e4188fec1171aee6626", "generation_seconds_batch": 16.47782673605252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.340320587158203, "genuineness": 4.030367851257324, "identity": 0.36592692136764526, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/21093ff0e36521c462835e1cf2adac27c092e2b0fef162eb3f95295d511e8b52.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "210baa39fda81caae0d9228c9e079f9c1a89756b6c2adbd5f18255e2de30fff8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ecfdebcca6e1a6d050db0ffdfc5d68dbe27594ed21234d2961886fab92053a6", "generation_seconds_batch": 22.39556314703077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.213237762451172, "genuineness": 4.519899368286133, "identity": 0.6052061319351196, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/210baa39fda81caae0d9228c9e079f9c1a89756b6c2adbd5f18255e2de30fff8.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "210bec3946136887126f2831d7f7ffdca4701c46242137ae25537c65259f7760", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9a86848f8c0431ce26d9266d3d65c0e3df5ed59351e1d46a182cc26dddd1b75", "generation_seconds_batch": 22.80790967307985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 5.9431867599487305, "genuineness": 4.118232727050781, "identity": 0.5081416368484497, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/210bec3946136887126f2831d7f7ffdca4701c46242137ae25537c65259f7760.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "212fb49c15e4e7bdbb8e015bedae72deebabf63e3d1b4a072858c576bc25ea7c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41cafec776e2267424d2d8eb3be9b14ba65650e155e5be32c880630b2d94aa9e", "generation_seconds_batch": 23.423893323983066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.48480063676834106, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/212fb49c15e4e7bdbb8e015bedae72deebabf63e3d1b4a072858c576bc25ea7c.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "21456066ca15febf7adf66711eea96e7b1d7cfb4492b83d11574d3d797da53c6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5aceb0c60bd64ec5a0bf42db8a5b652534990621ecce9232710493aff1876359", "generation_seconds_batch": 22.681843991973437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.136737823486328, "target_z": 1.3203123921809754, "ce": 6.054384231567383, "genuineness": 4.149612903594971, "identity": 0.5252739787101746, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/21456066ca15febf7adf66711eea96e7b1d7cfb4492b83d11574d3d797da53c6.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "2179f0792bdf8fcadc917313a286a7e80bbd578764a220237b4cb348ee129905", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2603dad92d6259904ecedde7143ae0e47bc16a55e7dbfce212d2cf2fa154900b", "generation_seconds_batch": 16.895594342029653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 6.2767791748046875, "genuineness": 4.2751336097717285, "identity": 0.7423785924911499, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2179f0792bdf8fcadc917313a286a7e80bbd578764a220237b4cb348ee129905.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "21bcb891eb77a5fb4effbf68b5a70ab302c3dcfb73c186e8e6e3bd39ed1bf4ea", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a571e6287f5bbce80a006b9c3a0740cee9b8d316b120aeb74103ff427d2eb19d", "generation_seconds_batch": 16.703064661938697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.054384231567383, "genuineness": 4.406930446624756, "identity": 0.6244684457778931, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/21bcb891eb77a5fb4effbf68b5a70ab302c3dcfb73c186e8e6e3bd39ed1bf4ea.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "228beabdfb39f5025e2ccf633e8ce8f17295ff74fc86106defeb7cb40dab6c36", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ddcf0056504cd4d4552e77a706c25dd4da0ea9176461e7023dc07b5c49c7205", "generation_seconds_batch": 21.37740915908944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9464356899261475, "target_z": 1.1640624686483687, "ce": 6.229123115539551, "genuineness": 4.394378662109375, "identity": 0.4131045937538147, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/228beabdfb39f5025e2ccf633e8ce8f17295ff74fc86106defeb7cb40dab6c36.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "22b40bc992df59d59f1193784e9413305df897de5a5d15c2aa1bf4c8b1f787c1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "776edd1e9136cd72af83f64dc47ec127dcec4cbb39187bc274fc5a77608dd81b", "generation_seconds_batch": 23.184111732058227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.229123115539551, "genuineness": 3.7165660858154297, "identity": 0.33361777663230896, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/22b40bc992df59d59f1193784e9413305df897de5a5d15c2aa1bf4c8b1f787c1.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "231bc1d405047541b0856847137d09712f8dd01e3fa18ed55bec7f85626e0b9d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f92924e36e6d01d2dd9d67bcafec1224486cc56f80ab22ec84068398bfe1281", "generation_seconds_batch": 16.6901885290863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.348263263702393, "genuineness": 4.3504462242126465, "identity": 0.5119326710700989, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/231bc1d405047541b0856847137d09712f8dd01e3fa18ed55bec7f85626e0b9d.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "23cf1d8044d601885151c1f476a6b7ff45df1b35721851bf39f46b15d99e11e3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "869ce0fa203e30b604eab4b94499fefaed3b5f8408e1f6780d074f25c6c99e88", "generation_seconds_batch": 16.68164375203196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.054384231567383, "genuineness": 4.381826400756836, "identity": 0.7802189588546753, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/23cf1d8044d601885151c1f476a6b7ff45df1b35721851bf39f46b15d99e11e3.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "240e4e82883272f6ecde3c11c8f907e98a5257bd5e3cf873e46270999d85b798", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed0b9a727755d7146e59940cddf3fe85bde55762e4a8329ea7b9fe16f153b1d4", "generation_seconds_batch": 25.506835706997663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.308549880981445, "genuineness": 4.3504462242126465, "identity": 0.6949300765991211, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/240e4e82883272f6ecde3c11c8f907e98a5257bd5e3cf873e46270999d85b798.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "24efd5a2ca52324fdecef7475821e986e9555d501774992c9aefeaeb5ea79e76", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "501d38da20586622ef9211fd0c4607323f47cf220037b5ef5a256b0d2ba34f96", "generation_seconds_batch": 22.466275697923265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.109982967376709, "genuineness": 4.2751336097717285, "identity": 0.782523512840271, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/24efd5a2ca52324fdecef7475821e986e9555d501774992c9aefeaeb5ea79e76.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "2563beba5e69be7ee7739ce206523592901b69038b1af82dc3bec71199e0467d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47ba6b3a777863a677be91b7a5df396d7ba1a49066883b2645066b71304940e9", "generation_seconds_batch": 22.722327993018553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.6045842170715332, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2563beba5e69be7ee7739ce206523592901b69038b1af82dc3bec71199e0467d.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "25daab053916399a3f7585dd05b94a47d91f634965509d483661df7793f32a33", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb9bb69c3804a6709e98ce31aa4ac03c732276c697e99885eac5d7f9b3ba56ac", "generation_seconds_batch": 25.182307439041324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.324435234069824, "genuineness": 4.306513786315918, "identity": 0.4975304901599884, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/25daab053916399a3f7585dd05b94a47d91f634965509d483661df7793f32a33.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "25e2268082592790d5dc4c20adea9d016bf4d6e574d8e6410e1020100ba82945", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81a5b066cc91a309af76455c21b4d93cdd6f0effc34d17b949c5250067176f6c", "generation_seconds_batch": 23.36297590401955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 6.173524379730225, "genuineness": 4.18099308013916, "identity": 0.6024796962738037, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/25e2268082592790d5dc4c20adea9d016bf4d6e574d8e6410e1020100ba82945.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "2609fe3b3f8c0adfffe6f185420f2ef868e0ed05af1fb6ef216ddefd68ba3155", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ce8474f7beb50aa6e542b0afab159970ddbde058969b2acab2163da99d24cae", "generation_seconds_batch": 16.21571815793868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.316492557525635, "genuineness": 4.099404335021973, "identity": 0.5611599087715149, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2609fe3b3f8c0adfffe6f185420f2ef868e0ed05af1fb6ef216ddefd68ba3155.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "2688e2efbb741e8a4f445206caab368c001f2e0fc67c4b2d69c6cb031ddf879d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "830f6752f66c805acf53e7baec214cbcc8a7a5433b180ce4cecbc785c0e9bde6", "generation_seconds_batch": 27.828801591065712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6704975366592407, "target_z": 0.9375000305869573, "ce": 6.205295085906982, "genuineness": 4.2374773025512695, "identity": 0.3453483283519745, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2688e2efbb741e8a4f445206caab368c001f2e0fc67c4b2d69c6cb031ddf879d.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "269c3387e894f4420ed9420fbb9937ecec72110303a907c20086a6e609582575", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfed79885065ed24f6e0d26edbdff42a1ba13d76f7ed8aca4ccfaafc2503714b", "generation_seconds_batch": 21.737582201953046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.588653564453125, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent. Ug.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/269c3387e894f4420ed9420fbb9937ecec72110303a907c20086a6e609582575.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "26c48282f3c7a1ae0d005a0e1f26ec04ab921f65329ebbfb50b47b6f1e6b5388", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90c91b3fc702339d438ce6c4938ed7432299db18095910cb12c997ee2162147a", "generation_seconds_batch": 17.917920239036903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.038498878479004, "genuineness": 3.8420867919921875, "identity": 0.0842687338590622, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/26c48282f3c7a1ae0d005a0e1f26ec04ab921f65329ebbfb50b47b6f1e6b5388.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc2cf62578610746cd6257dfba1fb6b3750ecb952dc29409ca7c9b64c15b22c9", "generation_seconds_batch": 16.912574254907668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.324435234069824, "genuineness": 4.18099308013916, "identity": 0.2987601161003113, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "26e7d53c0d6c956e05dc4005d6db0a442d3306700376fb967c0f94fec7535ce3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e398c3ebe6f5bea9d39c67fcbf93c2aec93ecd24fc65e3e460b73001dd1d6d80", "generation_seconds_batch": 17.012540786061436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9274054765701294, "target_z": 1.148437476295108, "ce": 6.3641486167907715, "genuineness": 3.8797430992126465, "identity": 0.28654855489730835, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/26e7d53c0d6c956e05dc4005d6db0a442d3306700376fb967c0f94fec7535ce3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "2730dfede23dfbf89109262a3a31f98c9249b00d16b79af4654bd3515fff076a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58d030b4c6415621687eec95a057785f7e28b3b783981c3740966f9b0ba64542", "generation_seconds_batch": 22.889224095037207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.181467056274414, "genuineness": 4.507347106933594, "identity": 0.7484008073806763, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2730dfede23dfbf89109262a3a31f98c9249b00d16b79af4654bd3515fff076a.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "273181d22fe8447eb2722dbdc601e8adf9e3a218432930f9c6debb6d2519c602", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bb616639fb830dfdec1dac38c4ceb78d4b3048649d2830ad67d2edf9640d474", "generation_seconds_batch": 23.855075791012496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.205295085906982, "genuineness": 4.143336772918701, "identity": 0.6558918356895447, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/273181d22fe8447eb2722dbdc601e8adf9e3a218432930f9c6debb6d2519c602.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "277fd60872ef524f5ae05e260afc8dbc1fccce07052c3d36b218d39ea42044c2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d30a7dc0ee4ec504b6a03c8477633bb3c40cfdedae33e1f5b3c24202565139f8", "generation_seconds_batch": 28.490188400028273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5087406635284424, "target_z": 0.80468754664511, "ce": 6.443574905395508, "genuineness": 4.432034492492676, "identity": 0.5266541242599487, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/277fd60872ef524f5ae05e260afc8dbc1fccce07052c3d36b218d39ea42044c2.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "27961a2aa82cc4a34d666445badcfe16e71e25ad3b379cab1bc684f7450fc0fd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30547c9c7ff1fd8f3f94b0774ff674f8d8153580798bda542cbbfea9f5c6837a", "generation_seconds_batch": 21.44706320797559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.40316104888916, "target_z": 1.5390624808831517, "ce": 6.443574905395508, "genuineness": 4.168440818786621, "identity": 0.4708434045314789, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/27961a2aa82cc4a34d666445badcfe16e71e25ad3b379cab1bc684f7450fc0fd.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "27c148faf571a4e27826ea16d1cfb3db944956faac26fca9dee5cddff2fe2be0", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56131a90dce62645b108c332fb0ed3bca27807177427b40e5cb0476cedbcf736", "generation_seconds_batch": 15.836011463077739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8789167404174805, "target_z": 1.9296875833494587, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.6099398136138916, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/27c148faf571a4e27826ea16d1cfb3db944956faac26fca9dee5cddff2fe2be0.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "27f63bd82e37ce819723625133d00b1d131ee5de691665ca2e73f739d02e4540", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5901567f1752b668f7c5febf93eb45be685f90312545d7971ad8e793b1ee9b19", "generation_seconds_batch": 16.09994700306561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.332377910614014, "genuineness": 4.143336772918701, "identity": 0.39635127782821655, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/27f63bd82e37ce819723625133d00b1d131ee5de691665ca2e73f739d02e4540.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "28663ffd1838492ab233d434712d6a982063930705aed64c62d95f139c37c448", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "70ecf4aa36a5245ff7d576e00e18d3341030678a92f8a72571418857c30a2489", "generation_seconds_batch": 14.85869769891724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.832254409790039, "target_z": 1.0703125145288048, "ce": 6.443574905395508, "genuineness": 4.494794845581055, "identity": 0.6275694370269775, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/28663ffd1838492ab233d434712d6a982063930705aed64c62d95f139c37c448.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "288442c56b6855bac2a6aa60cddde96ef61578c6fdce1bde301f08d6132b9e49", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e71250f7404b149058171cc856db78a8775b1a123f04f48127c561e75e1a07f0", "generation_seconds_batch": 21.575078852009028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6895277500152588, "target_z": 0.9531250229402181, "ce": 6.141753673553467, "genuineness": 4.425758361816406, "identity": 0.6870242357254028, "hypothesis": "J'ai trouvé la petite boîte, elle était près, de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/288442c56b6855bac2a6aa60cddde96ef61578c6fdce1bde301f08d6132b9e49.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "28ceff27ba6e72c328d2298f0af096c9a35a4e0057f7ea63928d55a73fb722fb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50c29398456ab3e2448954924ea7649911bf6982ccecb47552d3224e88d1e6f2", "generation_seconds_batch": 24.72769280395005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3841309547424316, "target_z": 1.5234375864081544, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.7133023142814636, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/28ceff27ba6e72c328d2298f0af096c9a35a4e0057f7ea63928d55a73fb722fb.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "28e9363bdf4ee6cff4a6c454811238f145f76a8a862d8c8aa8a6ccf0d4c88f49", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "296a6d452e3bd6e51fa4c002ce0a676593b921aceb0b3b95df38e2243fe66fb2", "generation_seconds_batch": 27.159921295009553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 6.38003396987915, "genuineness": 4.356721878051758, "identity": 0.4658350348472595, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/28e9363bdf4ee6cff4a6c454811238f145f76a8a862d8c8aa8a6ccf0d4c88f49.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "291beaba46b8aa6d0c302142bebdcf9e566f09ae1664fc5d4dbb3076ceb3b4de", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f47e999df0ad09d6918efaafd4fb0c2da555e527cce78465422c01b763a26667", "generation_seconds_batch": 23.86672034591902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.24500846862793, "genuineness": 4.532451152801514, "identity": 0.5885438919067383, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/291beaba46b8aa6d0c302142bebdcf9e566f09ae1664fc5d4dbb3076ceb3b4de.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "2928c1711c47c35959a12260b17dd9d3c62a9d56366fb14cde717761e84b2aca", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4f7d206a1f25387f43113851b351fe960b712b81bcc0349ff5ab0217e4b9b3e", "generation_seconds_batch": 24.121734102023765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 6.3641486167907715, "genuineness": 3.961331605911255, "identity": 0.31006696820259094, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2928c1711c47c35959a12260b17dd9d3c62a9d56366fb14cde717761e84b2aca.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "293351957642ffb4a4a0cf6b3e5d9e8fb92321c313de36f30a029e15a40777f3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6dfc3fa116572e022e531a8e89f6b4dd281fb32dd81f7febb01dcc0374e563c2", "generation_seconds_batch": 14.411127874045633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.054384231567383, "genuineness": 4.2751336097717285, "identity": 0.5784683227539062, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/293351957642ffb4a4a0cf6b3e5d9e8fb92321c313de36f30a029e15a40777f3.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "29421eb43df11195a02e51ebc1e1ebe04be96d1987521f08bb0b6a411e3738be", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75af7cb6385cb77f865c4b8cdefe4297998a5ef32085e97b423a57415f645ec5", "generation_seconds_batch": 16.91844222601503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.40316104888916, "target_z": 1.5390624808831517, "ce": 6.403861999511719, "genuineness": 4.256305694580078, "identity": 0.32669150829315186, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/29421eb43df11195a02e51ebc1e1ebe04be96d1987521f08bb0b6a411e3738be.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "2970c7f04db254c539b6a6d2bf52d88668f32c02dbe9238050fd6e9b45558259", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4eb3ae32ba3c7f23c8d289c265dc92b29954f60a3491f9731608bf313245a09", "generation_seconds_batch": 16.801777162007056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.403861999511719, "genuineness": 4.281409740447998, "identity": 0.518471896648407, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2970c7f04db254c539b6a6d2bf52d88668f32c02dbe9238050fd6e9b45558259.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "29b20686cbf316b0ccd6e18936a4c2c96452eed8eb621d38d0a67cb7569b038f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "924e950a82ea1113eb8853993611a5c44c2d2f13cef66b372d4e20906ce82691", "generation_seconds_batch": 23.54422197304666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.41180419921875, "genuineness": 4.168440818786621, "identity": 0.49150124192237854, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/29b20686cbf316b0ccd6e18936a4c2c96452eed8eb621d38d0a67cb7569b038f.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "29c5d5c6045168a236c1af0fd89fa6c0683ff7ca818c5d6c1993ba960daa1ae6", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d95e40eeb4629ba53774238edb450c3c84c788edf56278fb5af522c0ad39c692", "generation_seconds_batch": 16.73273474094458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.136737823486328, "target_z": 1.3203123921809754, "ce": 6.308549880981445, "genuineness": 4.356721878051758, "identity": 0.5867024660110474, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/29c5d5c6045168a236c1af0fd89fa6c0683ff7ca818c5d6c1993ba960daa1ae6.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "2a3eef254ec20c73f60a44ab0a0abbc5b1da327f7c48ad395ba986568b89e950", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0eb3f3f3579077dd5626b63d4bdf367acaf223a7c2ca0803db78becce64f56bb", "generation_seconds_batch": 17.178703021025285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 3.751009702682495, "genuineness": 4.406930446624756, "identity": 0.6254302263259888, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2a3eef254ec20c73f60a44ab0a0abbc5b1da327f7c48ad395ba986568b89e950.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "2a749d55e591f762ec0ae0e75faa49eb55d001dbf578ebc72509beb033dc5c16", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79a1b0692869796cea57562f0fe54d456a72e16725165589cfe4e2d95d6c2413", "generation_seconds_batch": 15.687270550988615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5173425674438477, "target_z": 1.6328126307592428, "ce": 6.157639026641846, "genuineness": 4.331617832183838, "identity": 0.710056722164154, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2a749d55e591f762ec0ae0e75faa49eb55d001dbf578ebc72509beb033dc5c16.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "2a8d75381d298d6bfd35fa31bd78fcb448e38b3f4a7a7ac031939b1027c4a9b6", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e0fe9848aeff1e5602a8f4110783d5ba96f9f83af8e93869d58de77ed85df31", "generation_seconds_batch": 21.340411863988265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.165581703186035, "genuineness": 4.149612903594971, "identity": 0.7768919467926025, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2a8d75381d298d6bfd35fa31bd78fcb448e38b3f4a7a7ac031939b1027c4a9b6.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "2a8da33b128fb4e8e81f1cbf6f2fbf7e73babea559f9afb844b014bf945beb11", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0e203e07a7066217b294d1b542e85b9b056d319d72b22477c379f547c2dfa64", "generation_seconds_batch": 14.86330408800859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8703148365020752, "target_z": 1.1015624992353261, "ce": 6.3641486167907715, "genuineness": 4.381826400756836, "identity": 0.3287244439125061, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2a8da33b128fb4e8e81f1cbf6f2fbf7e73babea559f9afb844b014bf945beb11.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "2ad92a5c1577f368fa400cba28e5af74583820baadcc2c1aa8ce090a37885df7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a13e74963a22032af25afbcd4ddac4010e14ef0a0cbcc7fa85b39e9e9844e13c", "generation_seconds_batch": 15.832985684974119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.09409761428833, "genuineness": 4.105680465698242, "identity": 0.6804512739181519, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2ad92a5c1577f368fa400cba28e5af74583820baadcc2c1aa8ce090a37885df7.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "2b2cdfdf9fe1132b10d69cfbd0d77b28037272d2fbe1fd19e1901ac3371bfcab", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9b9e9341405b1cfbb5ecf0cc4e0c43dc15ffcd3c39061ea87ea7a048c31a2ac", "generation_seconds_batch": 23.121323812985793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.062326908111572, "genuineness": 4.256305694580078, "identity": 0.710939884185791, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2b2cdfdf9fe1132b10d69cfbd0d77b28037272d2fbe1fd19e1901ac3371bfcab.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "2b540a83239ec3f887b6845bac670b7fb9fb2cd7505ccc4620cc932266185ef4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "679de1e2ec328fc914598a6c181a46918c0d4d74897dc48bfeebae5530054458", "generation_seconds_batch": 24.6298312949948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.046441555023193, "genuineness": 4.18099308013916, "identity": 0.724143385887146, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2b540a83239ec3f887b6845bac670b7fb9fb2cd7505ccc4620cc932266185ef4.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "2b95037ff8749dbca1b12f8863f9a63e97eb85564c53934b1193282105fa023f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea5fb42702d53288d7f5b8ba0a94fa0ede5fe05e0af14b6241cab9ac64b5d1f4", "generation_seconds_batch": 21.690792967914604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.086154937744141, "genuineness": 4.293961524963379, "identity": 0.6794663071632385, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2b95037ff8749dbca1b12f8863f9a63e97eb85564c53934b1193282105fa023f.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "2bda280c2bae51c54b991fc365de342274988cf2c6ea406faea46f49061b94b5", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3fe577e4f959fa22a8c9c31ecda7783c28ac45f768cd070e042a340bdcf5983", "generation_seconds_batch": 17.013645151047967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.252951145172119, "genuineness": 4.337893962860107, "identity": 0.7443122863769531, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2bda280c2bae51c54b991fc365de342274988cf2c6ea406faea46f49061b94b5.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "2c044bab4a0f10c715d192b530f731c842a995c0f6caded434def945b0bf4c17", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d842756fc2f26cf46fa7ada8ff458bda82950608e2911d875cb0a716a13e82d", "generation_seconds_batch": 23.42861358600203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.7846066951751709, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2c044bab4a0f10c715d192b530f731c842a995c0f6caded434def945b0bf4c17.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "2c12940e2876d29780773fe185e4a47237e30117bcc3a541560cb079eeed3cb4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1e6b22d46d2cc27d0e80522877b5f379cd8246599cd0af108b09d9adcb43ade", "generation_seconds_batch": 22.247053981991485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.6767507195472717, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Nemmy.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2c12940e2876d29780773fe185e4a47237e30117bcc3a541560cb079eeed3cb4.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "2cdd955a7e2e53fd5159cde54fd60409e4f281d4eb36e517d5ee3633afb444a9", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "392c5e9dc82b3fe8300a88ca1deb6f356a90a93ff4eec7013f12127ecddaf9b9", "generation_seconds_batch": 15.934377996949479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.332377910614014, "genuineness": 4.256305694580078, "identity": 0.4794918894767761, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2cdd955a7e2e53fd5159cde54fd60409e4f281d4eb36e517d5ee3633afb444a9.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "2d03ada55c3be29512f45a4718b944f34c9e280d8b76b531714287fbbb5a6ac2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f93e14daa6e73f460b8e43b7ad121585cd8b4fab94152ab73c3208914012ca49", "generation_seconds_batch": 16.21385952900164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.332377910614014, "genuineness": 4.068024158477783, "identity": 0.4006505310535431, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2d03ada55c3be29512f45a4718b944f34c9e280d8b76b531714287fbbb5a6ac2.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "2d38b48c23e46bcc3313580abaf92ae30adb55b1af5d4537f3801260d2a6ad0a", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1801d3edfd4ca8dd65fba6b84adfc0ccf40abb793d8bb4947d632afcc2feb5e4", "generation_seconds_batch": 15.816915599047206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.149696350097656, "genuineness": 4.105680465698242, "identity": 0.45527806878089905, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2d38b48c23e46bcc3313580abaf92ae30adb55b1af5d4537f3801260d2a6ad0a.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "2d712e2d6dbb945dedb75f9b136981a58a5deb73a2be46a5aac4e9475dd2ffeb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d252acfad7744e72bc20e8936c8ef5a0bc6393590d25622600651662fc47bd5f", "generation_seconds_batch": 26.311546046985313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.3904060125350952, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2d712e2d6dbb945dedb75f9b136981a58a5deb73a2be46a5aac4e9475dd2ffeb.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "2deee521eed2ce4707569fb25cf18d07ff3fdb46ef5f4d5a7f3da32e00149184", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d31362a421e7b4b48bb8cd9a1af73a46a859fbe13ddc8d91936d66006a87a7b1", "generation_seconds_batch": 16.097556525026448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.819082498550415, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2deee521eed2ce4707569fb25cf18d07ff3fdb46ef5f4d5a7f3da32e00149184.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "2dfc7002e2c3e4e4679da8f30f0aeab27627b0b753ea5c035822c22c178481c4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d30285fcfc2885be1c48550312ff5d9765febfbbc00bbf45401c5d56067e9ee0", "generation_seconds_batch": 22.80790967307985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 5.796247482299805, "genuineness": 3.9801597595214844, "identity": 0.7948215007781982, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2dfc7002e2c3e4e4679da8f30f0aeab27627b0b753ea5c035822c22c178481c4.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "2ef08d8197f03c8a84ffefa7113324c3365d59fc48d449760612b0debf598ab3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49e45d743e26cc543f30faf5afdf350a043e962abdafbf4d91040ef543fcb1a4", "generation_seconds_batch": 14.85869769891724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.003526449203491, "target_z": 1.2109375435864143, "ce": 6.483288764953613, "genuineness": 4.469690799713135, "identity": 0.5323022603988647, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2ef08d8197f03c8a84ffefa7113324c3365d59fc48d449760612b0debf598ab3.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "2f507fe96f290b3c3ecb16e1dd1a5d0e1178a2ef2d3a84a903aff5b0e94c3f5e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90e12643f64891ce7355a55d56a928161cdf1d36abdf7f580e5b1146ae7778b6", "generation_seconds_batch": 23.12697563599795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.1020402908325195, "genuineness": 4.1998209953308105, "identity": 0.5081571340560913, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2f507fe96f290b3c3ecb16e1dd1a5d0e1178a2ef2d3a84a903aff5b0e94c3f5e.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "2fa6247beea38f876d043b3308c1e361a8e5402fa5f521056f7f5b73182a652f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6bd4e723f8bc8721cd180ae2218a42b1ff039e2c8b30075ce02988559608586", "generation_seconds_batch": 26.24695725599304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 6.395919322967529, "genuineness": 3.8420867919921875, "identity": 0.35341978073120117, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2fa6247beea38f876d043b3308c1e361a8e5402fa5f521056f7f5b73182a652f.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "2fb10fbc001b7cda1c7ba5bec76c6deb02a77089ac22b678987a4b1c073ef58f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b175c0393c6727e13e488e6656267f6fa4bfb853086f8e78921349d6b8b4ddf", "generation_seconds_batch": 17.70455317304004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 6.395919322967529, "genuineness": 4.331617832183838, "identity": 0.0055225929245352745, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2fb10fbc001b7cda1c7ba5bec76c6deb02a77089ac22b678987a4b1c073ef58f.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "2fce402fdd9e94dd2db59d5eb0730ba8cc43f1db274297e37c846f742271828e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07b0a4c9a76dbe723a31f93f45795c414df4418417cf20572f276bc260fd191a", "generation_seconds_batch": 15.695993277942762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2119455337524414, "target_z": 2.2031249984706522, "ce": 6.300607204437256, "genuineness": 4.111956596374512, "identity": 0.7263766527175903, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/2fce402fdd9e94dd2db59d5eb0730ba8cc43f1db274297e37c846f742271828e.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "302e8355733d6507396f5e1d858c66e7c1f5ce8ef889a49d8209952c604d8123", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd68998e3dea80332795f1704cc3d18d51024493955568ac6dcadaeaeddee4d7", "generation_seconds_batch": 23.754654710995965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.205295085906982, "genuineness": 4.1998209953308105, "identity": 0.6632252335548401, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/302e8355733d6507396f5e1d858c66e7c1f5ce8ef889a49d8209952c604d8123.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "307f99a6b7f47d758e6cdd0c9c4d4e61464019caf17efb2d7fe051ed74d8ca82", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "692cf40489785b052045bdce4a78d30ff0074f7247b08f4a2d04f97ec10a9a22", "generation_seconds_batch": 16.27183519001119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 5.998785495758057, "genuineness": 4.105680465698242, "identity": 0.6561033725738525, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/307f99a6b7f47d758e6cdd0c9c4d4e61464019caf17efb2d7fe051ed74d8ca82.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "3081981e1858f5a9d2db3251fce679e15368ceca8277ad6dec40913ec6cf3b5b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e925265cc3b935424be34c25ce186b92dc5cae58e5af6ea2bdcc5726ddbf4544", "generation_seconds_batch": 22.67541510099545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.125868320465088, "genuineness": 4.024092197418213, "identity": 0.4305185079574585, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3081981e1858f5a9d2db3251fce679e15368ceca8277ad6dec40913ec6cf3b5b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "3092d5cd37cb6aaeeb602884651897744143f34f6aafb4782e0b84516e8d9b82", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a6704d71625c13107b64f154c87e2b589d3700a42503a1f31a9a7c45dd748cd", "generation_seconds_batch": 15.687270550988615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.802795648574829, "target_z": 1.867187418179889, "ce": 6.046441555023193, "genuineness": 4.293961524963379, "identity": 0.47370582818984985, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3092d5cd37cb6aaeeb602884651897744143f34f6aafb4782e0b84516e8d9b82.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "311c778143a6d29e3c34d63ca2741e9092d1d92c4959d3755a2c8d3afcac71e8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc44d2c00a60e29b2558735d607bc98f2a06d3896c2a3589f41767a2ad882782", "generation_seconds_batch": 27.57249904202763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.252951145172119, "genuineness": 4.325342178344727, "identity": 0.45042622089385986, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/311c778143a6d29e3c34d63ca2741e9092d1d92c4959d3755a2c8d3afcac71e8.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "316b80ca1acbae07f3e4052ded330fad7820276759412af2397a4dea1ae27913", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a30add44696b3326c4c31abf83d9473499e7bb62759bd8cf2f461345cf22f097", "generation_seconds_batch": 22.968566749012098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.054384231567383, "genuineness": 3.8797430992126465, "identity": 0.4424578547477722, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/316b80ca1acbae07f3e4052ded330fad7820276759412af2397a4dea1ae27913.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "3185758e32e4686d460491e392e1f9e7e7d902aac4a9e48162a7ab5345f49678", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d37f94ce70e7df7626797e3a78e893112efb71e5a17f690469ab32321ea24e0", "generation_seconds_batch": 22.53786337794736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.006728172302246, "genuineness": 4.256305694580078, "identity": 0.588646650314331, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3185758e32e4686d460491e392e1f9e7e7d902aac4a9e48162a7ab5345f49678.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "31ac6c57cb5a68b6f6f21941ceeed64b210462c2cf048c159373be2d92b2aac1", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6b93be1e7ebd05a822b47b0a7c173041830051ab0872ea2749e30c07bb12fd1", "generation_seconds_batch": 16.772040053037927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6505539417266846, "target_z": 1.7421874793538037, "ce": 6.332377910614014, "genuineness": 4.362998008728027, "identity": 0.6468921303749084, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/31ac6c57cb5a68b6f6f21941ceeed64b210462c2cf048c159373be2d92b2aac1.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "31d4d310879a42353a2e66c4fee6836f2a18aa65540ed6493654301e8a56dbec", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9534df5e99ade40833af0c9b4a2a8fde7671b8634ecf4b3b792b3730fefb7952", "generation_seconds_batch": 22.124390727025457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 6.24500846862793, "genuineness": 4.3881025314331055, "identity": 0.4604944884777069, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/31d4d310879a42353a2e66c4fee6836f2a18aa65540ed6493654301e8a56dbec.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "321c38bc692cf85a3fce1893410dc18d543d0f35a17d3fb22fc967ab208c79c6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5491cb300bce9b2a31f82d50690c37f206bac2e31c53b37a6bd27f3252df3bb7", "generation_seconds_batch": 16.852653230074793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3841309547424316, "target_z": 1.5234375864081544, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.5675152540206909, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/321c38bc692cf85a3fce1893410dc18d543d0f35a17d3fb22fc967ab208c79c6.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "323e85213ad159c73d2b5242be0bd6121ee8dc76030d1f7ec6778e78b0cba1f6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b75248e4a2a07c5712fffd7b39e4a6381e404ce9b85b6400913516cddd67ec23", "generation_seconds_batch": 25.688200602075085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 5.49045467376709, "genuineness": 3.8671910762786865, "identity": 0.1990164816379547, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/323e85213ad159c73d2b5242be0bd6121ee8dc76030d1f7ec6778e78b0cba1f6.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "32aa88e9832fcceb3a3f0f42baed8e4e03349df69e8fe2f8625b250d576a10e5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afe1635573e1141050cca14080d795fab576c764c73197a6b5cad5a6f9749f2b", "generation_seconds_batch": 17.098034358932637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.308549880981445, "genuineness": 4.369274139404297, "identity": 0.5454806685447693, "hypothesis": "De duur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/32aa88e9832fcceb3a3f0f42baed8e4e03349df69e8fe2f8625b250d576a10e5.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "32bd672b1a80cc69700cd3be77952233f07bad931308e2616e60539a20498ef5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a2fb22a9f732c6c2f9734c3eb3bd23e2aaaaf38f17eb85a1aea0249ed50e959", "generation_seconds_batch": 22.365785385016352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.149696350097656, "genuineness": 3.9425034523010254, "identity": 0.47166305780410767, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/32bd672b1a80cc69700cd3be77952233f07bad931308e2616e60539a20498ef5.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "33406576c9bb96ad9e582b5b4276a3710706831895e225c099ab08e3cf300f91", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58936e11a19738837e2aeac2f3c9c90ae7ca56dc5a6f987d8de99fe69c7f52d6", "generation_seconds_batch": 24.809131453977898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.0146708488464355, "genuineness": 3.823258876800537, "identity": 0.6481198668479919, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/33406576c9bb96ad9e582b5b4276a3710706831895e225c099ab08e3cf300f91.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "33d70d69eeb9e87cb3b6a0befcfc5bae7f01c10f78ffc17260c61732a1d69def", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2805ab88a3a4db8564d5155c21d1004974a28d6f7a77ce28919bfe4d5a1bb961", "generation_seconds_batch": 15.444292893982492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6877009868621826, "target_z": 2.5937499051804322, "ce": 6.213237762451172, "genuineness": 4.005263805389404, "identity": 0.6615540385246277, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/33d70d69eeb9e87cb3b6a0befcfc5bae7f01c10f78ffc17260c61732a1d69def.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "33de82cd4cee095ff4d5bba37391673183f345eb58cabde142828779db22dc70", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e25e5fb822b75ef82aef5dc38a89515387c85667e2936e527bdd138c3df47bb2", "generation_seconds_batch": 15.856424267985858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8037090301513672, "target_z": 1.046874977059782, "ce": 5.879645347595215, "genuineness": 4.607763767242432, "identity": 0.7451661825180054, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/33de82cd4cee095ff4d5bba37391673183f345eb58cabde142828779db22dc70.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0045ea7896c4c3b35bdf5b53cd1937aecc8a14aa8edf7d6b1f38e14167e97f1a", "generation_seconds_batch": 15.321692676981911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070132255554199, "target_z": 1.2656250657619583, "ce": 6.260893821716309, "genuineness": 4.3127899169921875, "identity": 0.7254970073699951, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "343549867d21f9b7b9a45cc7eb285aa36fc1791844da3a37ee42e73dea5bdde3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c178f48f7815bf2fc982c148c1c651969e0c8d776d2a8e24c34592c5ce98e519", "generation_seconds_batch": 15.856424267985858, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.0146708488464355, "genuineness": 4.268857479095459, "identity": 0.6551574468612671, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/343549867d21f9b7b9a45cc7eb285aa36fc1791844da3a37ee42e73dea5bdde3.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "343943ecac4763c3cc0730464cf94eb4e7c84091f35735bff6ca6b1f931413cb", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3bdfa42d905d602768bb2fd1bf32647da72b71a601d5317521d9ee6be9132eb", "generation_seconds_batch": 16.31849658500869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.173524379730225, "genuineness": 3.7793264389038086, "identity": 0.2503882646560669, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/343943ecac4763c3cc0730464cf94eb4e7c84091f35735bff6ca6b1f931413cb.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "346715d5f02b9c97ca026211e6839296a9b9b8162a68c17bb4232633eb1389bd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2e70e9009102fbe4061a5cc405ee98f3c81814d5c5b2143475e2918975a7baf", "generation_seconds_batch": 22.681843991973437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9844961166381836, "target_z": 1.19531245335489, "ce": 6.078212261199951, "genuineness": 4.545003414154053, "identity": 0.6651467084884644, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/346715d5f02b9c97ca026211e6839296a9b9b8162a68c17bb4232633eb1389bd.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70861058c94534c463e842a08544c225bf10952204012ddd26771de90b25db3b", "generation_seconds_batch": 17.281980382977054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.022556781768799, "target_z": 1.2265626338179385, "ce": 6.435632705688477, "genuineness": 4.20609712600708, "identity": 0.3690180778503418, "hypothesis": "The road turns left, behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "34fe5607d270d8507d4a6018a190a8305b58175e678e35fa500fb864b508940f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab211b39a5536c3dd36a8d18ba6a87d15e29d1c81126e705074b7fd207952a52", "generation_seconds_batch": 24.10219299700111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.173524379730225, "genuineness": 4.193545341491699, "identity": 0.727284848690033, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/34fe5607d270d8507d4a6018a190a8305b58175e678e35fa500fb864b508940f.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "352f76bea85a76627304c3b3337b3a73f96569a36432ac146a7757566fbaccf2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a82c7208e54a1a2389833e875b229a30936729219bced295bf3dbd5e621e223", "generation_seconds_batch": 16.821244201040827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.165581703186035, "genuineness": 3.760498285293579, "identity": 0.4141601324081421, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/352f76bea85a76627304c3b3337b3a73f96569a36432ac146a7757566fbaccf2.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "353646bfe7bd57f813b0c91de8fa5b51b07763fe4b23f020ce796cb34a207ed4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d39ebc754bfc711d0e6792aa8bfa794424ea086496b4db37ec141946e969a1c", "generation_seconds_batch": 23.12579206400551, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.292664527893066, "genuineness": 4.645420074462891, "identity": 0.788867712020874, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/353646bfe7bd57f813b0c91de8fa5b51b07763fe4b23f020ce796cb34a207ed4.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "357f86d5a8e0b292b7c3fc0b31f49abd7a656c456e801c8ba815cca4f529e62e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b047fef79a3ea2c4eb76927199dfa860dc28101794e790d5702ca9d33aa3c90", "generation_seconds_batch": 23.27275126997847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3841309547424316, "target_z": 1.5234375864081544, "ce": 5.967014789581299, "genuineness": 4.143336772918701, "identity": 0.6193956136703491, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/357f86d5a8e0b292b7c3fc0b31f49abd7a656c456e801c8ba815cca4f529e62e.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "359b934194c04b80d8cd8dee4714447ae202b27c9969eb3eb48f72623a2992b3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f19216743d309890e30b08cbb70bc9c18d7892b8746505c65c104829d008a947", "generation_seconds_batch": 16.872943460009992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.300607204437256, "genuineness": 4.243753433227539, "identity": 0.48200523853302, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/359b934194c04b80d8cd8dee4714447ae202b27c9969eb3eb48f72623a2992b3.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "35ac59d93d9eb3ebc974de3ec8d37f9e88ad47ca0a096abcbbcc773335ed3d9f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba744dd95cd97323701f59c91ea0b71b40c03b33355281807aad4de3d5f97b23", "generation_seconds_batch": 23.43190095096361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9464356899261475, "target_z": 1.1640624686483687, "ce": 6.292664527893066, "genuineness": 4.394378662109375, "identity": 0.49353551864624023, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/35ac59d93d9eb3ebc974de3ec8d37f9e88ad47ca0a096abcbbcc773335ed3d9f.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "35d433e71849f4ea8c5f43cb1dad01d100ec6a8c1dce72f5baf31c20a0bd78ce", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "955ebe5c33b667f64191a1b072e9ceac11cb52fe8e51ee8872f663170cad7b0f", "generation_seconds_batch": 16.932238746085204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 5.967014789581299, "genuineness": 4.168440818786621, "identity": 0.7528373003005981, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/35d433e71849f4ea8c5f43cb1dad01d100ec6a8c1dce72f5baf31c20a0bd78ce.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "366da3b8d6eba4aab547f616ab2068df8a4a29ed72b844a4e9df5137c1a6b416", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23c9db106580e3667c5dcd06607e5b7ee011d69826ea1f42c8c44cbb183a5a10", "generation_seconds_batch": 23.62145049299579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.4219162166118622, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/366da3b8d6eba4aab547f616ab2068df8a4a29ed72b844a4e9df5137c1a6b416.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3752184a06b6642e3dfa2219fab65f6a843f2f266f244a8fcc95442268cf8cdb", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06659d06517694be686d7b7abca59f6bef45e6ad2c7758b7c47c0de4bcdd0501", "generation_seconds_batch": 16.750297656049952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 5.959072113037109, "genuineness": 4.055472373962402, "identity": 0.567418098449707, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3752184a06b6642e3dfa2219fab65f6a843f2f266f244a8fcc95442268cf8cdb.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "376d9092e7a297a724137fca0d14cec86f6a151269766ebff3bd7aab3710bda2", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "139bf64e2eccaabd364c619465ac93a2352d0232df90d1afa688d0e892462504", "generation_seconds_batch": 17.543659736053087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.125868320465088, "genuineness": 3.9676074981689453, "identity": 0.6915842294692993, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/376d9092e7a297a724137fca0d14cec86f6a151269766ebff3bd7aab3710bda2.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "38cd10cff0ebcb29074ec5052babc132e183263cac77764f4e834caa67e343b7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2dbe60aa305a065e2e28b2589e8d01ca51e9e9b1b49652b72ed3007ac883136b", "generation_seconds_batch": 16.216055219992995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.1894097328186035, "genuineness": 4.15588903427124, "identity": 0.7062510848045349, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/38cd10cff0ebcb29074ec5052babc132e183263cac77764f4e834caa67e343b7.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "3957b7a65bedd78ec7743ba0965eab61cb18fa31a1cb7a85430985a3b8a10289", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "237638181871a20a404cbc1daa6db30a3fdce60558557fcefbc0c0233f8f8f53", "generation_seconds_batch": 27.39713989198208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.7531588673591614, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3957b7a65bedd78ec7743ba0965eab61cb18fa31a1cb7a85430985a3b8a10289.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "39892a84eb2cc56c2818eaa9117d98fddc6239e2ba52b116e3d92400ef03ee24", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57750ec611497dd8afcd91cd311e8fdca3eb38b34c4f66e5d238902db53b2fe1", "generation_seconds_batch": 17.886718077003025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.49831223487854, "target_z": 1.6171875405277185, "ce": 6.173524379730225, "genuineness": 3.7416701316833496, "identity": 0.38604170083999634, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/39892a84eb2cc56c2818eaa9117d98fddc6239e2ba52b116e3d92400ef03ee24.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "399257a78019de1e7508d71c7d68f66ad114098c535310cad6cb71c7ba027ba4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5275a8d80816818e71914345702d42bed52feb48e1bf520196671901659e40c0", "generation_seconds_batch": 26.63357648998499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 5.974957466125488, "genuineness": 4.055472373962402, "identity": 0.7122130990028381, "hypothesis": "La route tourna gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/399257a78019de1e7508d71c7d68f66ad114098c535310cad6cb71c7ba027ba4.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "39b1a7ef50ae73cb686ba1fd863ff0fd18ca79c7fa79c02aefdc07a027d59ab5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eac401f4b2a5ce09a0685668ecb89342e8cb55ffba66fb26b365f5795c9f4c84", "generation_seconds_batch": 25.266535451984964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5744330883026123, "target_z": 1.679687509940761, "ce": 6.221180438995361, "genuineness": 4.356721878051758, "identity": 0.578224241733551, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/39b1a7ef50ae73cb686ba1fd863ff0fd18ca79c7fa79c02aefdc07a027d59ab5.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "39b36b93c5ec9810576c6a77177336f856e97f8310999aebfeaa5733335be6c1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47fb6a77f01abe7453478977b1a13fd05359773d25c061c3e4f6c0d6d0df8aba", "generation_seconds_batch": 16.6739911820041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3080101013183594, "target_z": 1.4609376169951118, "ce": 6.23706579208374, "genuineness": 4.356721878051758, "identity": 0.5251056551933289, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/39b36b93c5ec9810576c6a77177336f856e97f8310999aebfeaa5733335be6c1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "39d6de74184a19b70d7084283b94b9b47918e4b9e0b16176bfe2e5f7aab9cd94", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e475bce076b628d5bce1bce39bf7a5278a0af170554182300d555de5d56bfafc", "generation_seconds_batch": 21.29445360705722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.181467056274414, "genuineness": 4.331617832183838, "identity": 0.7201247811317444, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/39d6de74184a19b70d7084283b94b9b47918e4b9e0b16176bfe2e5f7aab9cd94.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "39e2d44b2bb7a7182897c6f9566c97506d80fa51397eda02941a6c10c32ae4e8", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c312ab47d80d8f29063a34bc78b82d7d0935d900794e4f00e2efd486888ab6e", "generation_seconds_batch": 15.902370056021027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 5.974957466125488, "genuineness": 4.124508857727051, "identity": 0.5351057648658752, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/39e2d44b2bb7a7182897c6f9566c97506d80fa51397eda02941a6c10c32ae4e8.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "3b43afa26098873cf0acbd0d8f01cd70a79840a76cb574900effbd4173adb1d0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c535ea4bfcd535d9ddee154c1f728b5d75ce0224048c0202e0c0ccdf6a570116", "generation_seconds_batch": 17.462935196002945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.09409761428833, "genuineness": 4.243753433227539, "identity": 0.743886411190033, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3b43afa26098873cf0acbd0d8f01cd70a79840a76cb574900effbd4173adb1d0.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "3b656ecb3188c86ce9632136bac7e22f2eb893cce94175b2fd185dc5d46e1f99", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "181280833023ceb69de09a0fdf62ac7d75078c4bbe264bda0086545ea7b4596e", "generation_seconds_batch": 16.851675446960144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.38003396987915, "genuineness": 4.457139015197754, "identity": 0.4837404787540436, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3b656ecb3188c86ce9632136bac7e22f2eb893cce94175b2fd185dc5d46e1f99.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "3b775c1f6331e2345dc686bfd8acd819837d9fc53127eec1a200d22030baa033", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e86a921e2f2a54792994e175de3390cbabb05f636ebc760f307e052123effc34", "generation_seconds_batch": 18.652273544925265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.046441555023193, "genuineness": 3.7416701316833496, "identity": 0.29046934843063354, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3b775c1f6331e2345dc686bfd8acd819837d9fc53127eec1a200d22030baa033.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "3ba029ecca8611b01839fc9a7df1772279f3f7112a349505d2eed6fd42a34768", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f28b40b3edcb79b254c3b7e30ad0f63575843e5a233878473e8f65b74b0b90a8", "generation_seconds_batch": 16.63327646604739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.356205940246582, "genuineness": 4.557555198669434, "identity": 0.5118309259414673, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3ba029ecca8611b01839fc9a7df1772279f3f7112a349505d2eed6fd42a34768.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c99bcbe17a8f9754cef22ec4b24e41ff41357efaa79563f520f7f856d3f8fabf", "generation_seconds_batch": 16.707111887983046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402247667312622, "target_z": 2.359374922003259, "ce": 6.197352409362793, "genuineness": 3.8671910762786865, "identity": 0.7562348246574402, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "3c66a00b110f4d65b010587dff9b811fe9ab7a66a0879c97efa940bd66a058aa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "342e13e619c34c5c55cb33fe24777398bbae495c7b899b5503db2ca950165b14", "generation_seconds_batch": 22.67541510099545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.435632705688477, "genuineness": 4.362998008728027, "identity": 0.5177904367446899, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3c66a00b110f4d65b010587dff9b811fe9ab7a66a0879c97efa940bd66a058aa.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3c6c62515dcf8548b46b018afd9fb9fc5f1080f59372abf5e02ed02b22029447", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b2f0b5f8f0cfafdf66fd7994919f1aa3de8dfb1fc7f9b89352d1f26de1d21c95", "generation_seconds_batch": 17.864783053984866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.054384231567383, "genuineness": 3.936227321624756, "identity": 0.44064438343048096, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3c6c62515dcf8548b46b018afd9fb9fc5f1080f59372abf5e02ed02b22029447.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3cce77fa30dd4778437dd6467353ef021972eb59b266dcc3b933f2f37312b77a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "102ed22af39d2ae7e890ca512e865f1a123d8656990afee8b707d716c553e66f", "generation_seconds_batch": 22.51104962802492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.936007261276245, "target_z": 1.9765624625309772, "ce": 5.824047088623047, "genuineness": 3.8044304847717285, "identity": 0.5757061243057251, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3cce77fa30dd4778437dd6467353ef021972eb59b266dcc3b933f2f37312b77a.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "3d3924e812efa5aaef35960832d3bde49936b92fb25ba85abcc23882f5e05188", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "809a04c64f9094b51e4aca7b7d60e060eeda16ddd9a86cffe10e3089e1a7b971", "generation_seconds_batch": 16.88952737499494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.252951145172119, "genuineness": 4.319066047668457, "identity": 0.7498483657836914, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3d3924e812efa5aaef35960832d3bde49936b92fb25ba85abcc23882f5e05188.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "3dbfee5a1b94d3e4562285ea173fa565928a4794da3253304046738966672065", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fff4db7806cc31c9b822e43628b02d068645145672457110373a41ea6a1b0852", "generation_seconds_batch": 19.55221467197407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.316492557525635, "genuineness": 4.036643981933594, "identity": 0.5785105228424072, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3dbfee5a1b94d3e4562285ea173fa565928a4794da3253304046738966672065.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3dc605d4b578f103f70ee71edf19f785b6ca763170b3dd09ad19f5b28ad7b9ab", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4636716304ba36a7a2b5803ea67d81b842743a2943cd2144efc7fae7628e6ed", "generation_seconds_batch": 18.467165970010683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 5.839932441711426, "genuineness": 3.8295347690582275, "identity": 0.8746016621589661, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3dc605d4b578f103f70ee71edf19f785b6ca763170b3dd09ad19f5b28ad7b9ab.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3e3829b49b0b5f08faff0767b7b180247c5f9f76382e772add0c63ff1a4df523", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27d38fc19dc09de232fda2412f8066710223468776712b7e224bbd0b71ccdb59", "generation_seconds_batch": 22.299937922041863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9550375938415527, "target_z": 1.9921875527625015, "ce": 5.959072113037109, "genuineness": 3.860914945602417, "identity": 0.7678415179252625, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3e3829b49b0b5f08faff0767b7b180247c5f9f76382e772add0c63ff1a4df523.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "3ea79fb810f06bb143b40ca013dfab41d3a93fbc5e82a73c6bb16aa9807079ca", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51b7483dd3e22cbb8602ae6f0a1f1ca7167833c3d43bc5be3b0c598e7367d0bf", "generation_seconds_batch": 25.78215732902754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.022613525390625, "genuineness": 4.218649387359619, "identity": 0.7561663389205933, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3ea79fb810f06bb143b40ca013dfab41d3a93fbc5e82a73c6bb16aa9807079ca.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "3f21a47917a7319c95b300c23bcc022051a417238d5468b11add88fcd9b415ff", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a339427d09b2d54d068d2979d2134c7cd2d34735706c87c21c5d24085c84a0e", "generation_seconds_batch": 16.126218472025357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 6.38797664642334, "genuineness": 4.193545341491699, "identity": 0.581079363822937, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f21a47917a7319c95b300c23bcc022051a417238d5468b11add88fcd9b415ff.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "3f321153400770f5f60c2f1c3da1ec6770ff3a9edc111f59e82acaad3df7ae9e", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcf879e518def25ae6c98ec4ab6090221d935e8f5b47af4bfab08ada683e9761", "generation_seconds_batch": 16.103974448982626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 6.268836498260498, "genuineness": 4.519899368286133, "identity": 0.862751841545105, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f321153400770f5f60c2f1c3da1ec6770ff3a9edc111f59e82acaad3df7ae9e.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "3f6dce56f43df8e7492041ec47eb7640ce984c207916285c911eda71c3f4a350", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61690dc9c2ba5ba508029b01f975166c48e6e26ec6e7d5ff2f0eaafe216a6803", "generation_seconds_batch": 22.74536431301385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.269949436187744, "target_z": 1.4296874365320635, "ce": 6.348263263702393, "genuineness": 4.394378662109375, "identity": 0.678444504737854, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f6dce56f43df8e7492041ec47eb7640ce984c207916285c911eda71c3f4a350.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "3f80caf03714528c102f894206c86c101332b82b6bed9d5f80388ffdacbf6ee1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a700e4dac6229acd7dd73833db12140c8b93909aa31167ff337edee37e30f0d7", "generation_seconds_batch": 21.499631618964486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.097764253616333, "target_z": 2.109375044351088, "ce": 6.070269584655762, "genuineness": 3.8797430992126465, "identity": 0.5335766077041626, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f80caf03714528c102f894206c86c101332b82b6bed9d5f80388ffdacbf6ee1.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "3f8ca40c4d3efd8033fb569d9d9b8bf1dc1db12df22cc04b383ea19c4fe8f577", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bfb68493cc5d513694df5ffe13890b45f1f526a3f7e88a099c52312384f156f5", "generation_seconds_batch": 17.769579211017117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3898017406463623, "target_z": 0.7070312710285331, "ce": 6.062326908111572, "genuineness": 4.720732688903809, "identity": 0.695257842540741, "hypothesis": "He encontrado la cajita. Estaba estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f8ca40c4d3efd8033fb569d9d9b8bf1dc1db12df22cc04b383ea19c4fe8f577.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "3f9d2333d3e5b2ae4e63624493228b89d539981195dd96724148a9c32164816c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf601ccb336c6e261edfee96a0345cf3b0f646f8b97d06b54fb93d87e274eeb0", "generation_seconds_batch": 15.98727486806456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 6.022613525390625, "genuineness": 4.093128204345703, "identity": 0.7127996683120728, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3f9d2333d3e5b2ae4e63624493228b89d539981195dd96724148a9c32164816c.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "3fb132c6b7b7b03463a3bb40001da1e11f8fe18ade7bb3ef83c81a304c29b043", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d6ba2c08688f70e293887bef7c2d872b2c4ea9ee624102bce928729f4313711", "generation_seconds_batch": 16.121033271076158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.3267916440963745, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3fb132c6b7b7b03463a3bb40001da1e11f8fe18ade7bb3ef83c81a304c29b043.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "3fb2a703cb3e183cd1ae362534d7d2bdc81b6a6d36005a6d9b7dc95de3c2bcac", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7094fbdc428ca91504837c5ad2ebb8ad54ace8ec5647e87bee712b4a0c80bfe0", "generation_seconds_batch": 23.546610806020908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.2767791748046875, "genuineness": 3.9550557136535645, "identity": 0.38440418243408203, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/3fb2a703cb3e183cd1ae362534d7d2bdc81b6a6d36005a6d9b7dc95de3c2bcac.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "400559bcb12aa4bfee844d155ff5de79367b64c9cf80b0a71fca2764eb25eeb7", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdeb21ecd2ccb07d4928a7654ba1bc58eec7af63cf0b05aafeabb146181016c0", "generation_seconds_batch": 16.34132209303789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.252951145172119, "genuineness": 4.080576419830322, "identity": 0.5151088237762451, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter, bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/400559bcb12aa4bfee844d155ff5de79367b64c9cf80b0a71fca2764eb25eeb7.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "40177e82798a7d7536620840596e7e133539c78e2767b007213881a6b6d666fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7285edb80c3a46837e8436042b46f3ef8cb89c841e6edfe9cd318e5cb5114306", "generation_seconds_batch": 21.20651698601432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3070967197418213, "target_z": 2.281250058115219, "ce": 6.205295085906982, "genuineness": 4.218649387359619, "identity": 0.5855479836463928, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/40177e82798a7d7536620840596e7e133539c78e2767b007213881a6b6d666fa.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "4065906bdb7aa7a3f8a6ba864883735064125d68e487b91caec10b9d1817f024", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7607fc6be8ec699a1cfb5f2be95de7652389d7a92d5a116e513945486ac41199", "generation_seconds_batch": 23.410440891981125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.205295085906982, "genuineness": 4.18726921081543, "identity": 0.7172679901123047, "hypothesis": "Escribí el número en él sobre, léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4065906bdb7aa7a3f8a6ba864883735064125d68e487b91caec10b9d1817f024.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "40b6522e6d3af0ea3c1fac9f116fe78839d777f4a3b3b3f9787a9586af3db095", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c7440b0fcd8df3654ef9e37e3f874b91e96ef8b5a77dc7cf694b07f6f1b01c2", "generation_seconds_batch": 21.58070328796748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.726674795150757, "target_z": 1.8046874487668465, "ce": 6.109982967376709, "genuineness": 4.105680465698242, "identity": 0.7729321122169495, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/40b6522e6d3af0ea3c1fac9f116fe78839d777f4a3b3b3f9787a9586af3db095.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "418401ae5c8021763efa84181bfa6e933bfcf89f06ae9d92b8cc4b3c7c31c3b8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e262b51a2741fef84d7e5e04a3006d10986a8db7334eaccb84b0dfc0ac7078d", "generation_seconds_batch": 25.002348406007513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.332377910614014, "genuineness": 3.7793264389038086, "identity": 0.29493147134780884, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/418401ae5c8021763efa84181bfa6e933bfcf89f06ae9d92b8cc4b3c7c31c3b8.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "4195f83b8590c54d1367f4878c6b7165ba15c739abe9526efe636707db83e7ad", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c48de91df2abea9420b8d2169ee1230136eb48e8b720fa7dc12bf59de879558", "generation_seconds_batch": 23.95931884809397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.802795648574829, "target_z": 1.867187418179889, "ce": 6.340320587158203, "genuineness": 4.243753433227539, "identity": 0.7443876266479492, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4195f83b8590c54d1367f4878c6b7165ba15c739abe9526efe636707db83e7ad.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "421173f6782b19250d2f9c1d93ab2dca613d082301e2c89c7d846c0c4d147c37", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a45e778b73bf8fd328705c36ada6765726a2d6fd7ad3fd8c82b0ffb929c4976", "generation_seconds_batch": 28.490188400028273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.660982370376587, "target_z": 0.9296874854711953, "ce": 6.4515180587768555, "genuineness": 4.406930446624756, "identity": 0.7410959005355835, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/421173f6782b19250d2f9c1d93ab2dca613d082301e2c89c7d846c0c4d147c37.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "42246db3a7f7469eb7bfe428db7df52d8cb0f9bfd9d161206632ebf813f9bdf0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62926c60236a08d4357c689906dfde8e9e7f945004d28b4434fd87a7edc46197", "generation_seconds_batch": 24.469419492059387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.165581703186035, "genuineness": 4.3127899169921875, "identity": 0.7009563446044922, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/42246db3a7f7469eb7bfe428db7df52d8cb0f9bfd9d161206632ebf813f9bdf0.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "4244bde971aa302aa85a513ec4610f8b6d18e4221e5930bad7b527969613eb3c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d48d77f4f319bf154033cc6b99ab8e5adadcee891104c85dba52879901dd3436", "generation_seconds_batch": 21.13964100810699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.348263263702393, "genuineness": 4.18099308013916, "identity": 0.4910745620727539, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4244bde971aa302aa85a513ec4610f8b6d18e4221e5930bad7b527969613eb3c.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "424e2f82f42f324ed95df5fac022e2b119bf971b0776110cc85c6001108792f8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f477f2fea0f9909d6f59fdf605167d1510b10e9ef6b905fa057b7b38536fe384", "generation_seconds_batch": 27.39713989198208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.7935453653335571, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/424e2f82f42f324ed95df5fac022e2b119bf971b0776110cc85c6001108792f8.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "424eae5ef184da358f769195f0c8ae111ba1cc375c34a97550d397f2fe3e5609", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "895eac878019d3bb7ad01d6b6f6bcdf89345b0a308a0957c5a54c8af84272859", "generation_seconds_batch": 16.09994700306561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.3641486167907715, "genuineness": 4.319066047668457, "identity": 0.49330630898475647, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/424eae5ef184da358f769195f0c8ae111ba1cc375c34a97550d397f2fe3e5609.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "42a012a5da413b0d9b5a6444b445a2e91f7bb5840017a1578542d296bf34c700", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5e6e4d6a822832ee868c31e033eac5b2efdf1bd05d6c32634d6e0403adab3d5", "generation_seconds_batch": 16.64721522293985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.24500846862793, "genuineness": 4.582659721374512, "identity": 0.4629126787185669, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/42a012a5da413b0d9b5a6444b445a2e91f7bb5840017a1578542d296bf34c700.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "42d1379e4f92c78a56926e6e6249ffe5ece2390bf0a9cc70d7a23cc9d1071f83", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13d7dd465fa27256512857c86987f1c519a07610101982b9f8171679a055f3bc", "generation_seconds_batch": 15.753662036033347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.038498878479004, "genuineness": 4.457139015197754, "identity": 0.627780556678772, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/42d1379e4f92c78a56926e6e6249ffe5ece2390bf0a9cc70d7a23cc9d1071f83.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "42e46a8a76b579e3a8982fe7a1196af3693cae6a1b0cf4388c840849c3bdc155", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d90247a1114069fbc1da03a45def7bbeb48888c48795ed7e8435e89af84079b", "generation_seconds_batch": 17.543659736053087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.832254409790039, "target_z": 1.0703125145288048, "ce": 5.982900142669678, "genuineness": 4.457139015197754, "identity": 0.5557525753974915, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/42e46a8a76b579e3a8982fe7a1196af3693cae6a1b0cf4388c840849c3bdc155.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "430cc6a47fed7d15a30be82f389a7dda4733a029fae32da71abf6421d7bcd3a6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca79d53e17f2c2d236af85e37f1a42dd5252b06b7f9b336ff86de1533f09e20f", "generation_seconds_batch": 22.712754824082367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3460702896118164, "target_z": 1.4921874059451061, "ce": 6.38003396987915, "genuineness": 4.168440818786621, "identity": 0.48581498861312866, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/430cc6a47fed7d15a30be82f389a7dda4733a029fae32da71abf6421d7bcd3a6.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "433379dfeb4257f070f5412a0f70b294987a2dfcfebad87ebdd357028f18d5ff", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "77b4d3fa28236467e4bb8822c5e8faa42da255a7074804c443e4589f10baf3db", "generation_seconds_batch": 23.678506216034293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 5.792276382446289, "genuineness": 3.785602569580078, "identity": 0.7299904823303223, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/433379dfeb4257f070f5412a0f70b294987a2dfcfebad87ebdd357028f18d5ff.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "4335396e0acf930455cd7f3c62eabb64c513bc8c083c8bcbead64958c3ec57af", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3a310b7d7b8c121a4222938a781271e4faf11f8ec137566e26426e71c5320e6", "generation_seconds_batch": 25.76499787310604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.141753673553467, "genuineness": 4.337893962860107, "identity": 0.5018719434738159, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4335396e0acf930455cd7f3c62eabb64c513bc8c083c8bcbead64958c3ec57af.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "43390b4e313ae12a038bd81fa84fb256d42ec347f5f4103424f64883a4d11547", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de381dd5330617655f3272a4f707f5ba7f2d48830d2ea5731562d2addcbacaa6", "generation_seconds_batch": 16.907245110021904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 6.2767791748046875, "genuineness": 3.7416701316833496, "identity": 0.5761940479278564, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/43390b4e313ae12a038bd81fa84fb256d42ec347f5f4103424f64883a4d11547.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cb073efca46df32d77052fed3d351a4d2394bdd0b55cc5554742402421489e7", "generation_seconds_batch": 15.321692676981911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.572857141494751, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "4360ed4e544e5667384198ff6940b627a87c7e319dbae45607ba1103cfd1c1c7", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15d3161ba67f3da122da1b4d049cb566d056f4ba456ab4f3fb7ee567430c882c", "generation_seconds_batch": 23.410440891981125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.0305562019348145, "genuineness": 4.218649387359619, "identity": 0.7310757637023926, "hypothesis": "Escribí el número en el sobre, léelo bien, y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4360ed4e544e5667384198ff6940b627a87c7e319dbae45607ba1103cfd1c1c7.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "4364e015d3fb5598bdb186694c738b123d3755c6f30d8c8401d91c446dfdaa74", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24c7788801eeb57f4ec9ca5584b1faa6ddc1c2589802dd9c9c97e0e2719ebe1f", "generation_seconds_batch": 25.358862838009372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.157639026641846, "genuineness": 4.218649387359619, "identity": 0.7687669992446899, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4364e015d3fb5598bdb186694c738b123d3755c6f30d8c8401d91c446dfdaa74.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d620c83c3608561bd04f937a4536888e5968322a44cec75aa2841f9318583222", "generation_seconds_batch": 16.792581256013364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.4760531485080719, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería A alta. La colocaremos después de comer.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "4427d83abdfbea7fa03a99016fab3a93fd820ae05e547bfa6d758775b24673af", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d0dc5b0c082ccb09423118999fcfbf9363be46cac257865eb693812fb507c03", "generation_seconds_batch": 17.32417581800837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.395919322967529, "genuineness": 4.18099308013916, "identity": 0.5881431102752686, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4427d83abdfbea7fa03a99016fab3a93fd820ae05e547bfa6d758775b24673af.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "443e9b296dd34f734889bb33c43b3403af39ea90203c8b0d1dd04290e98f26be", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b09e4c5c9d773b2d74a751b88e9d6adf6c18676d289fb5f4a8810ca5f4723f03", "generation_seconds_batch": 22.247053981991485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.141753673553467, "genuineness": 4.13078498840332, "identity": 0.7141350507736206, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/443e9b296dd34f734889bb33c43b3403af39ea90203c8b0d1dd04290e98f26be.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "4441654ad48e0c4f9df8c16f82afb68a3d214c680a4bc20845c08fd21766e163", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c12600e95a4bed9be7f687ce89a9786ef0caebc093336adf3d9611538cc5784d", "generation_seconds_batch": 22.585956795024686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.060616970062256, "target_z": 1.2578124227679326, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.7544634938240051, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4441654ad48e0c4f9df8c16f82afb68a3d214c680a4bc20845c08fd21766e163.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "446f874c89974e2e3d4a705780fa74701c1453dcf933a01a4e5fd75a073256c5", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "550bd367bd5e7c94554a7202c59a16b342d330951925a93501723a8c94e99a82", "generation_seconds_batch": 16.766077021020465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7837655544281006, "target_z": 1.851562523704892, "ce": 6.1894097328186035, "genuineness": 4.061748504638672, "identity": 0.71186363697052, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/446f874c89974e2e3d4a705780fa74701c1453dcf933a01a4e5fd75a073256c5.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "448b5e63f399a92626c19fb112b497fa685d6b2b2e45bde6f058840eb4b6a3d8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66687aed97e8465121690b95fb261891f23415da95e54ebbdc5b67cca6b54742", "generation_seconds_batch": 21.941196984960698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.135824680328369, "target_z": 2.1406250290576097, "ce": 6.213237762451172, "genuineness": 4.250029563903809, "identity": 0.7717621326446533, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/448b5e63f399a92626c19fb112b497fa685d6b2b2e45bde6f058840eb4b6a3d8.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "4544cae2f04b3114c09c5e2e7f7a2d7eb12e94f724a48137a0095b147162b59f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41d36489fb0644f55ddc07fc81d0377450199d8363b45b926dee5bb7b28656bf", "generation_seconds_batch": 26.03256964602042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.292664527893066, "genuineness": 3.9927117824554443, "identity": 0.25690388679504395, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4544cae2f04b3114c09c5e2e7f7a2d7eb12e94f724a48137a0095b147162b59f.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "4588f28deecb1d109fa3be3e9a39c429a0c1ba27f2fed9daea79472004c25ee1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b3536622e3a0c289ac8d2cc95f2d657d0326445d2279e6ad8eba0d018ff0cc4", "generation_seconds_batch": 17.6924574469449, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.109982967376709, "genuineness": 4.319066047668457, "identity": -0.1074359118938446, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4588f28deecb1d109fa3be3e9a39c429a0c1ba27f2fed9daea79472004c25ee1.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "45a9fe49a4da8a6e350da2efa126bd0c7f3727ebc604869ab7642a91bf753a06", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fa3c16191bf7a23a8d1b5c80534caecfd5b69f2a0b16fc740bc28558160419e", "generation_seconds_batch": 16.613817290985025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.078212261199951, "genuineness": 3.936227321624756, "identity": 0.7387943267822266, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/45a9fe49a4da8a6e350da2efa126bd0c7f3727ebc604869ab7642a91bf753a06.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "45cb9760c416c0d32b9e249ea4507fff2a3b96ee45e79d086cb4e9bf8e394b03", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c0014a8d54a729e2a276130bdf4279091c8203ab800b8dbb29ef497438ee687", "generation_seconds_batch": 16.21385952900164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.443574905395508, "genuineness": 4.017816066741943, "identity": 0.5642964839935303, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/45cb9760c416c0d32b9e249ea4507fff2a3b96ee45e79d086cb4e9bf8e394b03.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "171413435a22df7d12cd160b625e7946b5488f6f7f71aad7d8da239661d7ee61", "generation_seconds_batch": 16.830739223980345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.2767791748046875, "genuineness": 3.823258876800537, "identity": 0.6094443798065186, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "46b2de2ca799f961995e6011d9edfc51041cded41cf0420029e1bd45db46346f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e80d2f83f6da1cb008a0b204d60f99c1c30b195f617509b5d3db5f66b8965b8f", "generation_seconds_batch": 16.19473274005577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1738851070404053, "target_z": 2.1718750137641307, "ce": 5.95112943649292, "genuineness": 3.9425034523010254, "identity": 0.6000120639801025, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/46b2de2ca799f961995e6011d9edfc51041cded41cf0420029e1bd45db46346f.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "478fa38d40f386ce6cb0b7ce5d1ef999ef0b2f69dd63d19c2bf8d89802e2afce", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc6b39fe9bddd26b89b52a6646e9b4bbafdf78e924f6feb02b7bffc97b029c8d", "generation_seconds_batch": 23.78746170201339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.229123115539551, "genuineness": 4.218649387359619, "identity": 0.5140459537506104, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/478fa38d40f386ce6cb0b7ce5d1ef999ef0b2f69dd63d19c2bf8d89802e2afce.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "47a22a77c296b991cb3c29d9edd2a87dec7f83b2308c2cd523d28966cac8da62", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87f43fd9daf224e54d082fbece514af076ec812194c06fa3e26134a6133517ac", "generation_seconds_batch": 25.800869549042545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.726674795150757, "target_z": 1.8046874487668465, "ce": 6.2767791748046875, "genuineness": 4.331617832183838, "identity": 0.47094905376434326, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/47a22a77c296b991cb3c29d9edd2a87dec7f83b2308c2cd523d28966cac8da62.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e6a08a5faf217cceedb572627996b7f336be3d872a517299f145a8f4adeeb85", "generation_seconds_batch": 16.95426589192357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.6318550109863281, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "4817402d005023f7c178a854c14b95a458001054284d58016a88c4837dc0ae84", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17ad73858637aa870043960ec9d3a5f17aa3d6e753a2a545d93bf452dd34e892", "generation_seconds_batch": 28.460626468993723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.070269584655762, "genuineness": 4.331617832183838, "identity": 0.7130035161972046, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4817402d005023f7c178a854c14b95a458001054284d58016a88c4837dc0ae84.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "48506c229685b2657419e6a251f29255e6fdd870968dc94463ac6d36157ab64d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b323293851c84aba6c874343b114e6440dd0b545c1f351cf852fa04dcf6561f", "generation_seconds_batch": 21.737582201953046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3460702896118164, "target_z": 1.4921874059451061, "ce": 6.181467056274414, "genuineness": 4.124508857727051, "identity": 0.6540669202804565, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/48506c229685b2657419e6a251f29255e6fdd870968dc94463ac6d36157ab64d.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "488ceec6e1a182fafc08c87b8f542726498e8fc2f0b44c42eb888bdcf84d243b", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a73943342835504a4d1cc15a86e0a65e839ac45699a5e87e5aa2b76f9bb6bbe", "generation_seconds_batch": 17.012540786061436, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6038917303085327, "target_z": 0.8828125084114132, "ce": 6.459460258483887, "genuineness": 4.030367851257324, "identity": 0.2518636882305145, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/488ceec6e1a182fafc08c87b8f542726498e8fc2f0b44c42eb888bdcf84d243b.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8256012e9aea1441f04fff75fbff4f4a965837b2c625bf1db1678d5d3acd8b73", "generation_seconds_batch": 17.06627894495614, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.536372661590576, "target_z": 1.6484375252342398, "ce": 6.109982967376709, "genuineness": 3.9425034523010254, "identity": 0.5776748061180115, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "4a09ad4fc7a1b717e6da8eaed8d7c87076742ac9724b47696e7e309e1b879198", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad9e235f0d5e7a85334e22ee1f9da8ebb353a2c3d4deab330f16d33a73bcf14f", "generation_seconds_batch": 22.506809619022533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8789167404174805, "target_z": 1.9296875833494587, "ce": 6.316492557525635, "genuineness": 4.18099308013916, "identity": 0.5639314651489258, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4a09ad4fc7a1b717e6da8eaed8d7c87076742ac9724b47696e7e309e1b879198.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "4ab7ec2b4ecf03730af8934917f1dd889c4f3f596f67ad33f28a51115b4f4b4e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8f923c582cb98008e508e039518b155d547089f59bb4e03fcad55b566b1a0194", "generation_seconds_batch": 21.305803386028856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 5.998785495758057, "genuineness": 3.961331605911255, "identity": 0.5900958776473999, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4ab7ec2b4ecf03730af8934917f1dd889c4f3f596f67ad33f28a51115b4f4b4e.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "4b19dac4481566d679526840f6df2a0fdf0515ae7bee2ce5db89b51752d4b2c8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79ed007953428b522ab714a1e490b98f0fbc75964f0ecf00ade82fd4aa0384ec", "generation_seconds_batch": 26.24695725599304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8037090301513672, "target_z": 1.046874977059782, "ce": 6.284721851348877, "genuineness": 3.6789097785949707, "identity": 0.17111222445964813, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4b19dac4481566d679526840f6df2a0fdf0515ae7bee2ce5db89b51752d4b2c8.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "4b6573d8378ce76e0550abaf2cff813e1508b0ea5abd4826ff1dac9a79c26a69", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70275bfea849c7a64e26dae2e51c5ce7645d2e30c552e6fa215956c60b2670e2", "generation_seconds_batch": 23.11167255300097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 5.597680568695068, "genuineness": 4.6705241203308105, "identity": 0.6332126259803772, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4b6573d8378ce76e0550abaf2cff813e1508b0ea5abd4826ff1dac9a79c26a69.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "4b7d04ae71e1da15e334069000fee145ad6e0801202ef9f03a34f1295a3fbbfc", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf665fae08bb8e8218653c0d3069b1ace232807cbc780dfbe92339e479dd5407", "generation_seconds_batch": 26.486166815971956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.5038771629333496, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4b7d04ae71e1da15e334069000fee145ad6e0801202ef9f03a34f1295a3fbbfc.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "4bd19e6152755d07a0629334dedeecd5b37ccebf0be5459b7de303c16a644321", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9157acac9132607b3af5ad8d1a33c57122948f5416ea3f789d328cc358ea32db", "generation_seconds_batch": 25.27030546707101, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.3099594712257385, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4bd19e6152755d07a0629334dedeecd5b37ccebf0be5459b7de303c16a644321.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "4c7415e53bc32eea1a3f048a5e79a537d614ef05b37e210ceed30a762c80fcb6", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee967bc7c86d336e11dc7e6f31c58f4b9328a614517081d6270286923f7af160", "generation_seconds_batch": 16.73273474094458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.669584274291992, "target_z": 1.757812569585328, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.7127284407615662, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4c7415e53bc32eea1a3f048a5e79a537d614ef05b37e210ceed30a762c80fcb6.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "4c9787ae295fd981534d0169cfa9f601365042292d99416c8a6061209b4f25fc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8c4c604b0e897af53e56c32f2e5bd995bd55a068e495ffb08513438bc6018c3", "generation_seconds_batch": 23.581734286039136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.6603677272796631, "hypothesis": "He encontrado la cajita. Estaba junto a. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4c9787ae295fd981534d0169cfa9f601365042292d99416c8a6061209b4f25fc.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "4ca963f5189608b131a2b4f806f6adfc50ecb399e9ae863053d9f72a4d81a61d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbccbf68389d339b5fe16ebb1212b042131d6bbdc1aae90641b040bdf962bbe2", "generation_seconds_batch": 25.182307439041324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.269949436187744, "target_z": 1.4296874365320635, "ce": 6.1894097328186035, "genuineness": 4.118232727050781, "identity": 0.41879844665527344, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4ca963f5189608b131a2b4f806f6adfc50ecb399e9ae863053d9f72a4d81a61d.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "4ce7347373ac10cd41f47ea5a5fb78b2b2050b85a3e4e88499c3e15fe479b115", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9384a92ec09ccf1d5c3d3ae53f8f5664cd74d0d4151e3c19339511a4aa67aaa2", "generation_seconds_batch": 26.475559833925217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.284721851348877, "genuineness": 4.231201171875, "identity": 0.5335131883621216, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4ce7347373ac10cd41f47ea5a5fb78b2b2050b85a3e4e88499c3e15fe479b115.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "4e0c3ac5542a4bcaf44e969dd61fd0db971b2a17016063281900b2b00586860b", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e26f3a92fe8314534d4ba919d4e94ddbeeb854e5d95ce1134ff10c32047f011f", "generation_seconds_batch": 23.090415695915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.763822078704834, "target_z": 2.656250070350002, "ce": 6.173524379730225, "genuineness": 3.9801597595214844, "identity": 0.673984169960022, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4e0c3ac5542a4bcaf44e969dd61fd0db971b2a17016063281900b2b00586860b.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "4e86bea2226d7baebce0e5864b75e0b3054cd259e8586742782ef0681f55efc1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf44ba49b5978c53d4730f7686f46a22b52a29ccbdb90939cc101c1259ebf7d1", "generation_seconds_batch": 16.43689714395441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.078212261199951, "genuineness": 4.193545341491699, "identity": 0.7741695642471313, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4e86bea2226d7baebce0e5864b75e0b3054cd259e8586742782ef0681f55efc1.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "4eaff653bd1fce08e9cea1780b1818cf9b945b4fb74de3b32e8d23e386cce2ac", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a6b59aa8eed97eb1aead0abcc7fc7141cf938a3aa814b70c3187d89872e04b1", "generation_seconds_batch": 26.70070983504411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.395919322967529, "genuineness": 4.055472373962402, "identity": 0.2412634789943695, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4eaff653bd1fce08e9cea1780b1818cf9b945b4fb74de3b32e8d23e386cce2ac.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "4f1751cfdff47fce93396370ea148a183944a85ab584467bbccdf7d5e3ca2635", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70c8309b647726b92ffa8c972b930eb28d1fe3561a1cc448e999038e7e8ec847", "generation_seconds_batch": 15.29709417000413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.316492557525635, "genuineness": 4.055472373962402, "identity": 0.33197444677352905, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4f1751cfdff47fce93396370ea148a183944a85ab584467bbccdf7d5e3ca2635.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "4f9fd4ca4e0dd1b3929d6e80a0704d9b953bd09d58c64a717b238bebbf5b3d47", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af2c72c6128f93c3d0eabd7f8d5b7fdfcf0ccc6ecdd136c76ba227fc5397f752", "generation_seconds_batch": 16.48630112200044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.070269584655762, "genuineness": 4.231201171875, "identity": 0.7533739805221558, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4f9fd4ca4e0dd1b3929d6e80a0704d9b953bd09d58c64a717b238bebbf5b3d47.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "4fa194b50feda80dff87ec2369169d30602ec30f779e47d22059b2c9dd581910", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "289de198c755adb7cb7fdedae0270d11f918171e5ee91207e2e20335e4a274f3", "generation_seconds_batch": 22.459264909033664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.23706579208374, "genuineness": 4.287685871124268, "identity": 0.3158033490180969, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/4fa194b50feda80dff87ec2369169d30602ec30f779e47d22059b2c9dd581910.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "500506aee77b3317dbe4fb75e56d834aea98e5ae13dbe7080ee03f5db0f85be1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16b5433e286f52bcef686cb65975bf513558b30ded1ba1f6f90b43721ee8ccb5", "generation_seconds_batch": 22.920538848033175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.24500846862793, "genuineness": 4.432034492492676, "identity": 0.615507960319519, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken. Hehe.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/500506aee77b3317dbe4fb75e56d834aea98e5ae13dbe7080ee03f5db0f85be1.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "503aa9a11234fad74929a2fc48109953b0b2bc88e870e826e950c765d9ffe691", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94ebe346cc8aef67b64ef34bc6249d3acb720bec95426538d0ff9ab797c295f4", "generation_seconds_batch": 16.19473274005577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 5.855817794799805, "genuineness": 4.231201171875, "identity": 0.798011302947998, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/503aa9a11234fad74929a2fc48109953b0b2bc88e870e826e950c765d9ffe691.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "5046f8f5710174f39785ee3f6e3935fd3b5785e682f4cddaef23e8b03db804f4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f86e23be9c97ac570cedc390ed6478c0aecddff893ed9aaa9d1405af5f6c584d", "generation_seconds_batch": 26.81598642806057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.141753673553467, "genuineness": 4.532451152801514, "identity": 0.6272510290145874, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5046f8f5710174f39785ee3f6e3935fd3b5785e682f4cddaef23e8b03db804f4.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "51e62209b65dcfe70ba0ec9add05d423241cb62aa86437fa15a2b00c2cf1080a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28fdcd6c57d6bde9b8440b24fdfbe2510e1fbb52f35abc08b57c998c198a8b28", "generation_seconds_batch": 22.722327993018553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1177077293395996, "target_z": 1.3046874977059781, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.7012186050415039, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/51e62209b65dcfe70ba0ec9add05d423241cb62aa86437fa15a2b00c2cf1080a.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "522ba4fe0fbbe1d34771071fd25da4a036cdad9e32dcd2fe16a5dde430cb6f02", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8450cb01cff93c8ad3d0b2ca3ac8da2a76e716a88dd241612fa1d2936af9e36", "generation_seconds_batch": 22.174540085019544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 6.181467056274414, "genuineness": 3.785602569580078, "identity": 0.40834420919418335, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/522ba4fe0fbbe1d34771071fd25da4a036cdad9e32dcd2fe16a5dde430cb6f02.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "526b3ab1a0836adccb90dcaca8fc9fbe7a96309c6d9ce7684f5a62ee3a20e5ed", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba41a169e81a60c19e0038e411d93d394335f4ef1bee28266c66212ca3096aa4", "generation_seconds_batch": 22.565220220014453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.197352409362793, "genuineness": 4.036643981933594, "identity": 0.6872565746307373, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/526b3ab1a0836adccb90dcaca8fc9fbe7a96309c6d9ce7684f5a62ee3a20e5ed.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "52ed1d579ae67952dc4aa9cc300b59ce6691a18e27713ecfdd99232a909102de", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d911d7567164fbf6d838ff06dd0ace6d25f33e2454bc1bf9bf64116a574e500", "generation_seconds_batch": 23.805692305089906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.054384231567383, "genuineness": 4.256305694580078, "identity": 0.7046691179275513, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/52ed1d579ae67952dc4aa9cc300b59ce6691a18e27713ecfdd99232a909102de.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "533a380a3bd8e2178a712d7638fee24cfaa8326857a0f39a680bec870457f1e9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d0efbf53e146bf88bf2b5904663a253e61fa26b5765f70cfe29de13aaf80021", "generation_seconds_batch": 23.423893323983066, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3080101013183594, "target_z": 1.4609376169951118, "ce": 6.38003396987915, "genuineness": 4.381826400756836, "identity": 0.46152263879776, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/533a380a3bd8e2178a712d7638fee24cfaa8326857a0f39a680bec870457f1e9.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "5364f89af671249d18f2f0ed2fb67d633714c901f062c60df2c2c40fe3948248", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "beb3f0a394e30afd02f99ab69972e99c50b7c1a50eb4c9250b078e6557469dcb", "generation_seconds_batch": 16.872943460009992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8789167404174805, "target_z": 1.9296875833494587, "ce": 6.149696350097656, "genuineness": 4.3881025314331055, "identity": 0.7116907835006714, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5364f89af671249d18f2f0ed2fb67d633714c901f062c60df2c2c40fe3948248.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "53e78b7a3d74fe6512c375bf8817753fa988f11d67e936c6dd1a5a32c9761d0c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7b1b9cf96561f9fcfd2a503d84d9597cc35da446fd8f5b63f4b73bc84300efd", "generation_seconds_batch": 23.15567300398834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.372091293334961, "genuineness": 4.18099308013916, "identity": 0.32901448011398315, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/53e78b7a3d74fe6512c375bf8817753fa988f11d67e936c6dd1a5a32c9761d0c.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "5410ca690f374e43864960e71d2789c0d20ac98550614ed5d81c341fcdc9faea", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c5168ff63d1e7ab5a4fa382f7fd5a909ccee574afb8f1e5b2df632b7ef24565", "generation_seconds_batch": 25.907587827998213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5744330883026123, "target_z": 1.679687509940761, "ce": 5.839932441711426, "genuineness": 4.369274139404297, "identity": 0.5221907496452332, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5410ca690f374e43864960e71d2789c0d20ac98550614ed5d81c341fcdc9faea.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "543018abee66ce89ade23e5a6b39f9e72edcdfc0c2ea841a9abace015946e668", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57c7861f32a8750aed9b20dddcb8550e5d730e2f0e87079f7224b7a6c7584ecb", "generation_seconds_batch": 17.463426877977327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.395919322967529, "genuineness": 4.017816066741943, "identity": 0.5830127000808716, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/543018abee66ce89ade23e5a6b39f9e72edcdfc0c2ea841a9abace015946e668.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b714e76d0b348d787eb827f0d36538bb2e3a96b1ce892565b5864d8082ad9766", "generation_seconds_batch": 16.830739223980345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.340320587158203, "genuineness": 3.9801597595214844, "identity": 0.568663477897644, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "5485b5d4e74240b7aea89e65ef9ad3daa88f1b7b85f6d9f544c8e423cceec442", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2abd1c17e89b0bf75bd0796ba63f74046c54ae69e6b7ab1228db11d07c70aba7", "generation_seconds_batch": 23.034495058003813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.23706579208374, "genuineness": 4.507347106933594, "identity": 0.7999414205551147, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5485b5d4e74240b7aea89e65ef9ad3daa88f1b7b85f6d9f544c8e423cceec442.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "550b036fba348d091081e977249e7bb72ad6c913a5e207a9a61348f4712a935c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67e4cf931bddde5358f8c7869a105215ad93f6106555995dc7f4cfeb5641c85b", "generation_seconds_batch": 25.36704861500766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.284721851348877, "genuineness": 3.948779582977295, "identity": 0.28823965787887573, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/550b036fba348d091081e977249e7bb72ad6c913a5e207a9a61348f4712a935c.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "551647796905e7732c218610b26045614a5f80dff9ffad32b4150679fd2b2933", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4cc5366ce856234c7ff4f365cb23d169d93750850f1fd2a3ee4842443c1fa2e4", "generation_seconds_batch": 16.345214131986722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.109982967376709, "genuineness": 3.7165660858154297, "identity": 0.44237202405929565, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, liess sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/551647796905e7732c218610b26045614a5f80dff9ffad32b4150679fd2b2933.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "558840e1c4d098938a0e1df5d257d38cabd023d179178a06e78331bd1a84595d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0b90533a6e98946eb723c4ad7d2803881701fb951b9b0dac691212da93c3e41", "generation_seconds_batch": 22.565220220014453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.859886407852173, "target_z": 1.9140624931179346, "ce": 5.982900142669678, "genuineness": 4.243753433227539, "identity": 0.5591046810150146, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/558840e1c4d098938a0e1df5d257d38cabd023d179178a06e78331bd1a84595d.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "55ee784f61b9124706a766066c122de8bc22e27b47cbb578eb3ccc264bddbbfa", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c84c06c58afdf0d43d40f0de75c1953bb31b5bc641ed716e261bb8734df991d", "generation_seconds_batch": 18.662174213095568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9369206428527832, "target_z": 1.1562500214108702, "ce": 6.078212261199951, "genuineness": 4.256305694580078, "identity": 0.7631752490997314, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/55ee784f61b9124706a766066c122de8bc22e27b47cbb578eb3ccc264bddbbfa.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "5602eba8cdb22f556f1f810b37229be9e7adf1221efebe4196859d4914c014b7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53bea1aa7c9a7bbbfe812e50031d1e34e7420eabcb8672306f5e4075a33e5985", "generation_seconds_batch": 21.535951316007413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.2767791748046875, "genuineness": 4.137060642242432, "identity": 0.4777180254459381, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5602eba8cdb22f556f1f810b37229be9e7adf1221efebe4196859d4914c014b7.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "5603c7f2f8095583a48038d3d63e3def6d51fa600dca6d73e714a1bc8cf1ee30", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b063a8b7aff3b83567fd71265f0fa6dd1d54fbd79da935efc4eee7763cd40fcd", "generation_seconds_batch": 24.301791940932162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.24500846862793, "genuineness": 4.2374773025512695, "identity": 0.6022505760192871, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5603c7f2f8095583a48038d3d63e3def6d51fa600dca6d73e714a1bc8cf1ee30.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "56a8073b2fa445c6a42efd4878b8883c636af8ef31ddc9b3552dd80e640681bd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eade62a004341b5f2d7b61b86d4cef8422ff9fe6e57e3f6485e03a76d30b0a84", "generation_seconds_batch": 16.653242388973013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.165581703186035, "genuineness": 4.337893962860107, "identity": 0.4803787171840668, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/56a8073b2fa445c6a42efd4878b8883c636af8ef31ddc9b3552dd80e640681bd.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "579e707070331ec84ffe2e608aa48231743b7da31ea656dd5656edbaac81a010", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b4690a957a2b1fe884dc3aad60bb1b0f31e84e22a36161f66537f39ca98275c", "generation_seconds_batch": 16.15053148206789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 5.879645347595215, "genuineness": 4.381826400756836, "identity": 0.6464120149612427, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/579e707070331ec84ffe2e608aa48231743b7da31ea656dd5656edbaac81a010.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "57ca4955dc64e68fcf23b9abe531c0816f8b9a2a524819b5f0ce6db7f0a71dd4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5272bd31541f60f7fa05315c3fba8bcf4ddb9e9077962865ce1d9e8758ba0fe3", "generation_seconds_batch": 22.125607194029726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.324435234069824, "genuineness": 4.20609712600708, "identity": 0.5475156307220459, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/57ca4955dc64e68fcf23b9abe531c0816f8b9a2a524819b5f0ce6db7f0a71dd4.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "58ad822741eb9eee7b20decb020a72c91b8172e4fcba157ef81f681d0eb06e6a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e52ddaa4e0ed79a0f190be9e81949dc68b45c2b8b8da46036ca985b7493ffe78", "generation_seconds_batch": 17.33204422495328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.062326908111572, "genuineness": 4.406930446624756, "identity": 0.6064793467521667, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/58ad822741eb9eee7b20decb020a72c91b8172e4fcba157ef81f681d0eb06e6a.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "58c59f0b41ab7063872760f08b04d87b2ce8085c3461e9d172c3c4bc4a4d6b0f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ec8560d3177a27c4c6ef1c0e70bc022c12d58a27498029ae7b28b4c26db84ea", "generation_seconds_batch": 18.911488098907284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.1894097328186035, "genuineness": 4.3127899169921875, "identity": 0.7265708446502686, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/58c59f0b41ab7063872760f08b04d87b2ce8085c3461e9d172c3c4bc4a4d6b0f.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "592971080c32314cc42fa155a72a4e90f95c1c4510efa5847fed9d4af2c304ef", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe7ba5a20996e103d1398a977e1322111a12890ddf927e9c4a0f59ed66d1271f", "generation_seconds_batch": 15.74870394996833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.1894097328186035, "genuineness": 4.331617832183838, "identity": 0.6345552802085876, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/592971080c32314cc42fa155a72a4e90f95c1c4510efa5847fed9d4af2c304ef.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "598a5064cca291d6864694cd173d3ee04d3dd2c8f31c6dd84e95dd3635b5ace4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef08dc520367f2e1056a7eee9da7497cc520e825b727e56cb933a4ba9a0e73e1", "generation_seconds_batch": 27.828801591065712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.316492557525635, "genuineness": 4.21237325668335, "identity": 0.5731740593910217, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/598a5064cca291d6864694cd173d3ee04d3dd2c8f31c6dd84e95dd3635b5ace4.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "599f27d07ae73a6c750f7bc3493534126226a3d9fa0ff3d6c1f7327e689bd98f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7e79fad562463656833a78ddd26f57e669424eee5cfe72474534b12b6323639", "generation_seconds_batch": 23.67886245297268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1177077293395996, "target_z": 1.3046874977059781, "ce": 6.316492557525635, "genuineness": 4.18099308013916, "identity": 0.5041180849075317, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/599f27d07ae73a6c750f7bc3493534126226a3d9fa0ff3d6c1f7327e689bd98f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "59fdd1a0bfe087ff332de9878785db5b8a4fea35f96dff79f6692faa94b5ba54", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "700032749dfac7e77a367f50351504cee808b54f3326968972b502315f359753", "generation_seconds_batch": 23.805692305089906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.24500846862793, "genuineness": 4.118232727050781, "identity": 0.7031806707382202, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/59fdd1a0bfe087ff332de9878785db5b8a4fea35f96dff79f6692faa94b5ba54.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "5a3d7e6f1d2d94df2d15a7a40d3d9236c3c139d5b16f6666cdfe0bda1ce20895", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf7ca1990b33320cb37f5cc2a76b1c70a846f90fad0cc26ca7d3d88b00022837", "generation_seconds_batch": 24.091849363059737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.022613525390625, "genuineness": 4.2374773025512695, "identity": 0.8019945621490479, "hypothesis": "La porte est- encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5a3d7e6f1d2d94df2d15a7a40d3d9236c3c139d5b16f6666cdfe0bda1ce20895.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "5abccbecf008af11c632dc491bcfbb220424783b0b1c48151871e6915ab4ebdf", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a24cf881882e8e79b63ab420893cec0ba21ddeaf97a047c5bd0c4786e87ca947", "generation_seconds_batch": 23.782389735104516, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0796473026275635, "target_z": 1.2734375129994568, "ce": 6.403861999511719, "genuineness": 4.256305694580078, "identity": 0.418930321931839, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5abccbecf008af11c632dc491bcfbb220424783b0b1c48151871e6915ab4ebdf.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "5af363de0e75b696885aabf9bf56d5a0065946daa6ac7492923f991e7c912e41", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "85478055762dd3a670835f9410b6246df1b98cb40696f83d3f1e637eabe6b706", "generation_seconds_batch": 23.153994086897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.062326908111572, "genuineness": 4.105680465698242, "identity": 0.4369453191757202, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5af363de0e75b696885aabf9bf56d5a0065946daa6ac7492923f991e7c912e41.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "5af5cdd2f1c667c74bbb2e8860ff0e53112b17574fa0c86c6439a3796d46a67f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09f32992354a7b112032c02ca50fbec9ce067bc701e185ae1b3315e8ecb5451d", "generation_seconds_batch": 15.995794213027693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7180730104446411, "target_z": 0.9765624625309772, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.5299505591392517, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5af5cdd2f1c667c74bbb2e8860ff0e53112b17574fa0c86c6439a3796d46a67f.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "5b3d9e00e3378a7d9d43b9e864c5b78d6b4bad266604df95c3606b1da294f81d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14f0fcc1c351c8cf754960400a85381158efb6c78438148e4625e8131731dd0d", "generation_seconds_batch": 23.255556654999964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.23706579208374, "genuineness": 3.7165660858154297, "identity": 0.37862348556518555, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5b3d9e00e3378a7d9d43b9e864c5b78d6b4bad266604df95c3606b1da294f81d.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "5b73382d08b0859a0bce1c1aad78f0951cc2628b9708f5eeb46b1b957bae3915", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b07adfeea40203d69f78c76bef6f20c8b645f76b62e516cd9779cc302a2d889", "generation_seconds_batch": 18.218167974962853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.086154937744141, "genuineness": 4.218649387359619, "identity": 0.6224104166030884, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5b73382d08b0859a0bce1c1aad78f0951cc2628b9708f5eeb46b1b957bae3915.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "5b7dac2e252eeac2f13dc19ef2960544351bc5ff08d93beaed7f42847655482a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c998690f33b356e30ca7e280f6e80428cc1155210e4f96a7e568cf2a8bcb7f38", "generation_seconds_batch": 23.495719761936925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.669075608253479, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5b7dac2e252eeac2f13dc19ef2960544351bc5ff08d93beaed7f42847655482a.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "5c4cb04335629237a039937e30af5637f286001d45bed0cfbeb9658639401129", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e30d9bc027481f22c2eb77ab3f1125901227633d044c7fcc48b4deb827b39fd9", "generation_seconds_batch": 16.852517124963924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 5.935244083404541, "genuineness": 4.017816066741943, "identity": 0.6877717971801758, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5c4cb04335629237a039937e30af5637f286001d45bed0cfbeb9658639401129.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "5c546a9e508dfe2108f6d93ea69424bec423dbcf046080a7a5762bf0fe5a4bdf", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89c758f5a39e4e551b0c1bbe221e79ee5e8cf7e92bb1c800ec6f83703d41cc1d", "generation_seconds_batch": 14.185862428043038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.1020402908325195, "genuineness": 4.287685871124268, "identity": 0.7182924747467041, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je gewoon dat.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5c546a9e508dfe2108f6d93ea69424bec423dbcf046080a7a5762bf0fe5a4bdf.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "5c61e3683eecfa557e482d0e61f6d51db07150d5da7947c9897c0a94f8f753ab", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ea90a7ff9e7c76768baf1800f176476c5bb6d79e6f88ff5319c2af2d2a36e41", "generation_seconds_batch": 18.838251937995665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.5019407272338867, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5c61e3683eecfa557e482d0e61f6d51db07150d5da7947c9897c0a94f8f753ab.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "5c73ca123e6b8cbacc30848a37b72bd566a509b7d0f969e7abd61850b3094235", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "186a41cbeb97531e4f60e3dd1d6e40e0b874e7877e7483319c49d017c063930c", "generation_seconds_batch": 16.805085337953642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.229123115539551, "genuineness": 4.118232727050781, "identity": 0.7186522483825684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5c73ca123e6b8cbacc30848a37b72bd566a509b7d0f969e7abd61850b3094235.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "5c9e4ff29611b1473e3614f5fab716b45984eac4bf1034075a0bdc4c18a34e76", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c8405f7fc23df04b4b0524279ea2b0bd767c7dc7c621e4468669d883a789ea4", "generation_seconds_batch": 16.45454908697866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846788167953491, "target_z": 1.0312499847065213, "ce": 6.268836498260498, "genuineness": 4.632867813110352, "identity": 0.8078376054763794, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5c9e4ff29611b1473e3614f5fab716b45984eac4bf1034075a0bdc4c18a34e76.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "5cdb17ce4d297cd5e412caded7e9f5c68cf755f2c2caf2d74a1b0395520f943d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d09a45aa85538e1840e79ef4f39d94a8f4159a3f231b1ca95b004f1392d7d1f", "generation_seconds_batch": 23.838195130927488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.1894097328186035, "genuineness": 4.319066047668457, "identity": 0.7287018895149231, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5cdb17ce4d297cd5e412caded7e9f5c68cf755f2c2caf2d74a1b0395520f943d.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "5d2e0c0fa1a0b7fb208aa77cd3207fdf2692ada67e505c81d37d6d167f7c6e04", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be647efa074ae2bf4edde328d3429d7abc07aea61836be652dd129750513ca6d", "generation_seconds_batch": 17.078591127996333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.205295085906982, "genuineness": 4.369274139404297, "identity": 0.6659207940101624, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5d2e0c0fa1a0b7fb208aa77cd3207fdf2692ada67e505c81d37d6d167f7c6e04.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "5d65ad7f092181f57d1c5252e29ed8e65ae83d99c0a0faa2f192d7052fbdb4bb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48ae96a9007866d1743668f09f9d8d4784048aed273b48bbaf833b21ae711c67", "generation_seconds_batch": 21.181950255995616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 5.935244083404541, "genuineness": 3.923675537109375, "identity": 0.616669774055481, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5d65ad7f092181f57d1c5252e29ed8e65ae83d99c0a0faa2f192d7052fbdb4bb.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "5dee1d207580b7adf075d78a9c9ac07e0dbe0fd31a53285556f6e106f4bf7f5f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea7e5650847910bab9892fcd7200408eb7c39fda556eb942c4d4c163e522610a", "generation_seconds_batch": 25.506835706997663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.936007261276245, "target_z": 1.9765624625309772, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.6154100894927979, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5dee1d207580b7adf075d78a9c9ac07e0dbe0fd31a53285556f6e106f4bf7f5f.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31b669a8ce9e0f269c957e27a85c58c6f58a6f3ad08bff06817f8d2d5bcabde0", "generation_seconds_batch": 15.696578364004381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.7067599892616272, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "5e7470f711e2039d88aef23eb92bfa8bcc1a74fd4df63fd619d79504696b634f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c1a9227b1245de6d418fde8e1ac5b88df35103e0c3cdd527f059b92cec3f516", "generation_seconds_batch": 18.15141965902876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.09409761428833, "genuineness": 4.293961524963379, "identity": 0.47140103578567505, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5e7470f711e2039d88aef23eb92bfa8bcc1a74fd4df63fd619d79504696b634f.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "5e75033c525ef84c834e2f2b1789564b0a48cdb60757ca2306be9b1219118d82", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43556bbfa8d1f050270a693e33e6f999f9b10efcd1596e65866cca383c1bf7dd", "generation_seconds_batch": 17.08506784995552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 5.879645347595215, "genuineness": 4.093128204345703, "identity": 0.5628687143325806, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5e75033c525ef84c834e2f2b1789564b0a48cdb60757ca2306be9b1219118d82.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "5ef69996b18ea191860ea822586eed74b9d4153ed88bb3b7880e8ed7fc39831b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57e1caadbff799d9c0e9f2b46df59a49cfe45978c799889466c0b0466cee0989", "generation_seconds_batch": 24.668218436068855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193828582763672, "target_z": 1.367187467119021, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.5263434052467346, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5ef69996b18ea191860ea822586eed74b9d4153ed88bb3b7880e8ed7fc39831b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "5f19b4bc63d5bc5a66ef470cb3ea8fd5d3fec5984699610d0b611408191a3d5c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a3b70019a5b41ef0edc4838bed262912b0dbccfb4ad314bff55d1c956ccbf9d", "generation_seconds_batch": 16.907245110021904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.268836498260498, "genuineness": 4.432034492492676, "identity": 0.5061722993850708, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5f19b4bc63d5bc5a66ef470cb3ea8fd5d3fec5984699610d0b611408191a3d5c.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "5f4484bf7260c5ffacf4e6056dbc85b592601b27f4bef29cca4c7cf355371c35", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "234cda57184b73b9c14b81b31ad2470ee453a45bbb0f1d6382b81d57593c150b", "generation_seconds_batch": 23.12697563599795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.229123115539551, "genuineness": 4.362998008728027, "identity": 0.5669651031494141, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5f4484bf7260c5ffacf4e6056dbc85b592601b27f4bef29cca4c7cf355371c35.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "5f681283197b9deda25e5db4c0fbd74e7d0d054b130ece2dcf62beb272001d32", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f9e011ba1d8d90b8f5584cad845ee746eed289c80ab955a3d390f93238b68cc", "generation_seconds_batch": 15.438471636036411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.117925643920898, "genuineness": 4.356721878051758, "identity": 0.426726758480072, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5f681283197b9deda25e5db4c0fbd74e7d0d054b130ece2dcf62beb272001d32.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "5fdec407055a9aaf833865544f52a6e213bdc9b32356757d92012d0d82df5077", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "97cb314596629af26d3d571c1b1b3a8151351a0d4b63553424921f20530baddb", "generation_seconds_batch": 16.847032010904513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.197352409362793, "genuineness": 4.143336772918701, "identity": 0.8155790567398071, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/5fdec407055a9aaf833865544f52a6e213bdc9b32356757d92012d0d82df5077.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "604ffd6da36779949cb92976a81480887fbc0de04b1a5c28e5acaad0170cbdfe", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67273ee2bc67bb2cb70317b3eb7322d856bf6ba20ef91e25c4042637759ad9e5", "generation_seconds_batch": 17.20948301802855, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846788167953491, "target_z": 1.0312499847065213, "ce": 6.3641486167907715, "genuineness": 4.193545341491699, "identity": 0.36031532287597656, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/604ffd6da36779949cb92976a81480887fbc0de04b1a5c28e5acaad0170cbdfe.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "607a04617369dc25f67a02e1df13651fb175e6a0f4571db8e563bbc6a672614a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28b2a6998ea297ceb3468a9368a2a0e388f2909e43a8f7e2dafce8724d9a2515", "generation_seconds_batch": 18.838251937995665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.078212261199951, "genuineness": 4.262581825256348, "identity": 0.7372065186500549, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/607a04617369dc25f67a02e1df13651fb175e6a0f4571db8e563bbc6a672614a.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "60baf5d288fd8cd36a1c3313858ff6bb57b42186391bd4e7667d741e9d7d6b4a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5efc852518ddcc7701fe4c54713c41978d4e3f5816c88318b50527fc54b64ac3", "generation_seconds_batch": 16.772040053037927, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.348263263702393, "genuineness": 4.124508857727051, "identity": 0.7262860536575317, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/60baf5d288fd8cd36a1c3313858ff6bb57b42186391bd4e7667d741e9d7d6b4a.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "6139fc94900c25f9d02fda4972931a60e44764bf74d492a12d3a0b8b33b4d177", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44c8d6ba56af022f734a5df8192a8981ce1c9695f6e6518d84c7b95f2ebae2a7", "generation_seconds_batch": 17.254525818047114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.7441892623901367, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6139fc94900c25f9d02fda4972931a60e44764bf74d492a12d3a0b8b33b4d177.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "61d033d4d364f573015fee344101eaacb2c4493295dc43f58bd6a69ed406bc8d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb8be489955038b1e061c7e92e991c7f6f1e0466db79c8cc6797aab90d824a74", "generation_seconds_batch": 23.36297590401955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.141753673553467, "genuineness": 4.293961524963379, "identity": 0.7995408773422241, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/61d033d4d364f573015fee344101eaacb2c4493295dc43f58bd6a69ed406bc8d.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "62105aea6d19ade7d957fbc984d6f8072c1d006ea8f8574049729a7d2244a9b9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5eb11e8d2576e4afa12db80013c24cd26382180eae6053518dfeae492091834", "generation_seconds_batch": 16.00133712904062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0415868759155273, "target_z": 1.2421875282929355, "ce": 6.4515180587768555, "genuineness": 4.482243061065674, "identity": 0.39842814207077026, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/62105aea6d19ade7d957fbc984d6f8072c1d006ea8f8574049729a7d2244a9b9.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "628fe3d1f6c6fbd7608528ed683d5806d2051c1f2f5b4f63524a4f9f9b16e630", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f324d505da41c1aa87a565e79d94cdb51febea0d9ec09c1e026c455fbaba2230", "generation_seconds_batch": 19.46838003094308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.16047091782093048, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/628fe3d1f6c6fbd7608528ed683d5806d2051c1f2f5b4f63524a4f9f9b16e630.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "62ea2c52b6902f7b60e7f879f8b59d5e4fd05f3ed384b9f01a6dc231d6eed98b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8eccd54acf7bb395e5d6cf957274ecd8408d16fe7139d150043656672106fda1", "generation_seconds_batch": 22.506809619022533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.221180438995361, "genuineness": 4.174716949462891, "identity": 0.751401424407959, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/62ea2c52b6902f7b60e7f879f8b59d5e4fd05f3ed384b9f01a6dc231d6eed98b.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "6343e1110eee59da8946aef35bf1e2aa0852dd6c9a7de2d6d4ce170580ab683b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "160f496f4764d701b6f7b35477a6fa196ced71c1c70dc5fd5c7dcda41c1e72da", "generation_seconds_batch": 16.12785638589412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 5.95112943649292, "genuineness": 4.287685871124268, "identity": 0.7226499915122986, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6343e1110eee59da8946aef35bf1e2aa0852dd6c9a7de2d6d4ce170580ab683b.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "6363165100ed0afb4dd8eb7e3cd7b9b0afff7eed40bf29f04bf52956846191b8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c46feab57cd3d0670d680fb362c77c2aba107700e1436c906080bf1c0030a66b", "generation_seconds_batch": 26.70070983504411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.324435234069824, "genuineness": 3.8797430992126465, "identity": 0.5704663991928101, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6363165100ed0afb4dd8eb7e3cd7b9b0afff7eed40bf29f04bf52956846191b8.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "63adcd3625b40008864ddad12d8f0c20776e7450ca088f5f76be0f255b046277", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31b5e05cb9eebcefeed34da311c1d3b99255f714e183a969c80ac3dd70d1f126", "generation_seconds_batch": 21.9562211119337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.300607204437256, "genuineness": 4.482243061065674, "identity": 0.6078245639801025, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/63adcd3625b40008864ddad12d8f0c20776e7450ca088f5f76be0f255b046277.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "63ae4c7e22d2e362bea75267270c815dbf9470f67b2ea83694ab3be3f211f8db", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a16110e65c0a367dedae0690cd61e220c0566f3d2934763af7beb7b5ed3b4cdc", "generation_seconds_batch": 17.70455317304004, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.097764253616333, "target_z": 2.109375044351088, "ce": 5.585766792297363, "genuineness": 3.8671910762786865, "identity": 0.21600216627120972, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/63ae4c7e22d2e362bea75267270c815dbf9470f67b2ea83694ab3be3f211f8db.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "63c80926a426a8ce7804500ea62ecff452ab24a466bd09a6a986beeb9cbc6a6f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15273918781a706d27cc449434bf9c9c727e318d4854e4bb34d393a26ef353d0", "generation_seconds_batch": 16.59505917003844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.149696350097656, "genuineness": 4.6203155517578125, "identity": 0.8274385929107666, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/63c80926a426a8ce7804500ea62ecff452ab24a466bd09a6a986beeb9cbc6a6f.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "64242f39c884503d4bbee8d9475606837a0a71aa2eec3bef5873edef470e4935", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b27effe3b9e2c023ba219188418e6a0ec41aaaa3756655dc69892f7228a7c594", "generation_seconds_batch": 24.1676687739091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3555855751037598, "target_z": 1.5000000489391319, "ce": 5.736677646636963, "genuineness": 4.055472373962402, "identity": 0.7217050790786743, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/64242f39c884503d4bbee8d9475606837a0a71aa2eec3bef5873edef470e4935.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "6564196c2b2407f038b8d69dd9daa27d2bb3ea7779a3c90943f6574951a1ab7e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff21d535d4558c591d9646d2caf76416e3e6baf8f1aa2624c127455d669057b0", "generation_seconds_batch": 22.51298378407955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.205295085906982, "genuineness": 3.9927117824554443, "identity": 0.16523036360740662, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6564196c2b2407f038b8d69dd9daa27d2bb3ea7779a3c90943f6574951a1ab7e.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "65906b5fe6a9af923cd773b6aafd1a81bd873bf11f5fd4729d740c62f174a0e6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fe7e1743b09126711115536a9f10efcb08e755cfc31e7cc46cd0d45d279ddc5", "generation_seconds_batch": 23.184111732058227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.141753673553467, "genuineness": 3.6663577556610107, "identity": 0.25737008452415466, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/65906b5fe6a9af923cd773b6aafd1a81bd873bf11f5fd4729d740c62f174a0e6.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "65adb5878f879d9f8ca248cbaea0835abe6b99f9ed1cc575db94b1798941bc4a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4808eda2dc77e8f10d2d4f632abcbfafec8daec8ffafc28e31b8b2545989524c", "generation_seconds_batch": 24.72769280395005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.165581703186035, "genuineness": 4.444586753845215, "identity": 0.5911538600921631, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/65adb5878f879d9f8ca248cbaea0835abe6b99f9ed1cc575db94b1798941bc4a.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "65f1e16d9bfbf67517492fc4b99f06f72028855ad39343b7cc56043d7017a720", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21f082286623b0427672597cbd17f903bc06d0b53c1697d66fbfc3dc3d77841d", "generation_seconds_batch": 16.971699091023766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.125868320465088, "genuineness": 3.7165660858154297, "identity": 0.6741164922714233, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/65f1e16d9bfbf67517492fc4b99f06f72028855ad39343b7cc56043d7017a720.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "6607cd6544c250d18e3e7785e555b5520c6f05fab9db50b22c45a63ec961f977", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f382725ea88005e8efd474cef11cb1deec71a5d473aff2cedcdfae7a9c5268cc", "generation_seconds_batch": 22.570028182002716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.0146708488464355, "genuineness": 4.457139015197754, "identity": 0.7731175422668457, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6607cd6544c250d18e3e7785e555b5520c6f05fab9db50b22c45a63ec961f977.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "66ce70819fc46ad84dbb42cd4773a50cb2f25670824b6c70a8fac092b815141f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59ff132d7630ca43fc27e16b0089fbe49c632ed1e0a5091fede4a53a7e15b0fc", "generation_seconds_batch": 24.45141274004709, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1747984886169434, "target_z": 1.3515625726440237, "ce": 6.125868320465088, "genuineness": 4.406930446624756, "identity": 0.7375391721725464, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Éloi Limerne.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/66ce70819fc46ad84dbb42cd4773a50cb2f25670824b6c70a8fac092b815141f.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "67a20a2eaf6f144fa80e5982352708d3e356f49d97c9010fba9b6e3b27a9dae1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "250834d4a218c0d38638aef431d9ca3a1018ce53520331a38c439f51a1078b01", "generation_seconds_batch": 16.80598330590874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.7228802442550659, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67a20a2eaf6f144fa80e5982352708d3e356f49d97c9010fba9b6e3b27a9dae1.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "67b134c58e760a49579f7d00b88522c9de6590631442989ff332ceede25e882b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9995dc6332f4268ed8eaff905f4fe6bc1fcb9bc87ea2583e2ee5c2c68330bb97", "generation_seconds_batch": 21.317286453093402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 5.967014789581299, "genuineness": 4.218649387359619, "identity": 0.6517527103424072, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67b134c58e760a49579f7d00b88522c9de6590631442989ff332ceede25e882b.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "67c67b6f34de4f37233dab39b4637620712dbe817aff3e807e7eeb235cb248e7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4f236ab577c3e33a19da057a5d2ed155edfc9b93d677aa4f98f2534135b37045", "generation_seconds_batch": 22.74536431301385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8607996702194214, "target_z": 1.0937499541195639, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.22224974632263184, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67c67b6f34de4f37233dab39b4637620712dbe817aff3e807e7eeb235cb248e7.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "67db0d98ac874a46e8426a30ad33ca1cdf0db993953facf804a01ad4793c48dd", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41b8d9a5da8da4b0a5fed5156625fd22fb8698a50403a638d6024ef5bb2e06d4", "generation_seconds_batch": 23.675143119995482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 6.205295085906982, "genuineness": 4.042920112609863, "identity": 0.6271430850028992, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67db0d98ac874a46e8426a30ad33ca1cdf0db993953facf804a01ad4793c48dd.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "67e3494d62ea1162e951c4837b5315336a50400143145f16952e9f09c6b916e9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c62d21c0acffe04976e018fa89cb8c2172a5cafcfcd4fb6be1adf7de4f5392f8", "generation_seconds_batch": 16.45454908697866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.086154937744141, "genuineness": 4.080576419830322, "identity": 0.6088273525238037, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67e3494d62ea1162e951c4837b5315336a50400143145f16952e9f09c6b916e9.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "67e693a0a9fa91302db289715b1b50a1c1d7817cf40b5f8373d2a7b8439fa057", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae68c97d5007b39314f2c8a8d9cd1e2e7435d1eb87998868958e1088b11d2062", "generation_seconds_batch": 23.62145049299579, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.260893821716309, "genuineness": 4.281409740447998, "identity": 0.5816595554351807, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67e693a0a9fa91302db289715b1b50a1c1d7817cf40b5f8373d2a7b8439fa057.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "67fce6cccd4a110b5540f209e9f6d807dd66e43b3dc2371eb947080cb366776f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afe838d9179bfb1a2f0ea94109d0166299cde84ead0aa6a6b0d1226d43e6c58c", "generation_seconds_batch": 16.805085337953642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.324435234069824, "genuineness": 4.394378662109375, "identity": 0.8079309463500977, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd: zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67fce6cccd4a110b5540f209e9f6d807dd66e43b3dc2371eb947080cb366776f.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "67ff794c04b1eb8b062a3f735e0a4fb3a7664010e0d127d913911c89bd0b456e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4cd516b908ccb6388634c80e6b67ea867a08c76b7619403f96ed5e4495b13d2", "generation_seconds_batch": 25.678539952030405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.6872231960296631, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/67ff794c04b1eb8b062a3f735e0a4fb3a7664010e0d127d913911c89bd0b456e.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "68565e34079ad6420a91820e13e50c84e799fd1e7e5bdcc2ae2aef2a4cb375fe", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b9e97107e97af1f1b49db8c86fa7e481382ef16ae78b89a086747eb408f24d3", "generation_seconds_batch": 15.574978390941396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.38003396987915, "genuineness": 4.017816066741943, "identity": 0.7125351428985596, "hypothesis": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/68565e34079ad6420a91820e13e50c84e799fd1e7e5bdcc2ae2aef2a4cb375fe.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "688d5422c9d242fa81fad448c8483175c4da2fe54540157f3128f1c4498f94fd", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f72fd0426b498440408ab1622b370252a9295182ac138d35e08a86ce82f79587", "generation_seconds_batch": 18.585814098012634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.38003396987915, "genuineness": 4.381826400756836, "identity": 0.3713111877441406, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/688d5422c9d242fa81fad448c8483175c4da2fe54540157f3128f1c4498f94fd.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "68be0f2e3732b47eadd7d63adf41a83674148efe68bdf63643af89f3ccdfc553", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e15cd5276435a22362c04ae24b0ca7d80d84f67310928bb9f17a4987f818560b", "generation_seconds_batch": 21.29445360705722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.6347968578338623, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Ah ?", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/68be0f2e3732b47eadd7d63adf41a83674148efe68bdf63643af89f3ccdfc553.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "68ed62ab05a4408fd0672d423369ded05cf8206b66624a103f5ee09feb3b75e5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc1fd515fca14c3cd3bd22c7fe7c512361c13fc8ec4bf5c40d4a26659783869d", "generation_seconds_batch": 20.887658805935644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.252951145172119, "genuineness": 4.243753433227539, "identity": 0.7346504926681519, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/68ed62ab05a4408fd0672d423369ded05cf8206b66624a103f5ee09feb3b75e5.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "69b83aede956bf49efa63db1f46d8465b811d72cd8cb69eda9d54992e19d0060", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3caaf26589f5e3255c6f2e4061a36b02bbf765b66baba0986f966fb4b8f3a2d", "generation_seconds_batch": 20.91312189400196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.4515180587768555, "genuineness": 4.005263805389404, "identity": 0.22616814076900482, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/69b83aede956bf49efa63db1f46d8465b811d72cd8cb69eda9d54992e19d0060.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "69c4a59c7168bd77984c0110bd7e52608e73b2afe397e0ac2fc4a2b234e6d7a2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0485adcbaa64f29d04ae0a46426d90760e3370463dd4409ae7dc93a7adad78c4", "generation_seconds_batch": 25.866032514953986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8218259811401367, "target_z": 1.8828125084114133, "ce": 5.959072113037109, "genuineness": 3.961331605911255, "identity": 0.6803604960441589, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/69c4a59c7168bd77984c0110bd7e52608e73b2afe397e0ac2fc4a2b234e6d7a2.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "6a591fc9b06488f444e35c55328a1c2dda973e29dda39f1c86ad10a2a8c1e8b8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02cd79f500bdb134c2cfcdc408815aa927db89d4acc05a2cb5b479be4fa82dd8", "generation_seconds_batch": 22.842010893044062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.2767791748046875, "genuineness": 4.337893962860107, "identity": 0.6510928869247437, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6a591fc9b06488f444e35c55328a1c2dda973e29dda39f1c86ad10a2a8c1e8b8.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f72b12cdc958aa52b2b215e3f631e3f44e7bb002c4aa00e7f554664df7e88a3", "generation_seconds_batch": 17.013154405052774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.23706579208374, "genuineness": 3.923675537109375, "identity": 0.3013855218887329, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "6aeb9dc811e3fea2eb3df8932d7af0b03dbc7c6734430c4e1cf206739ecd3d50", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1220f06be489f22632a7afee4de9d0ae8ebe87280b164ca1deec61b0390ac0ca", "generation_seconds_batch": 23.576122549013235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 6.09409761428833, "genuineness": 4.287685871124268, "identity": 0.714228093624115, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6aeb9dc811e3fea2eb3df8932d7af0b03dbc7c6734430c4e1cf206739ecd3d50.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "6af8e67922f69f03c60ae5861ea4c2c049da671d8deb1082d2d4c02e09df2839", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "76af8227f54157af322bdfc5aab4e8d36105e9eb96c94a31749b69df64f6561e", "generation_seconds_batch": 20.16824153799098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.8433626890182495, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6af8e67922f69f03c60ae5861ea4c2c049da671d8deb1082d2d4c02e09df2839.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "6b11767880a7cbacd02c17b488031d03f535e2a9a596a13b0266d21df9cbf932", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a38bf3bd2de24cbe7bf87eccdf981d9cbcca5aa892f195f32cf2f1127cc5271a", "generation_seconds_batch": 17.6924574469449, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.022613525390625, "genuineness": 4.20609712600708, "identity": 0.6273372173309326, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6b11767880a7cbacd02c17b488031d03f535e2a9a596a13b0266d21df9cbf932.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "6b5c154f01c98126828b5ba8f592337cd1f03388fcb5d322255e029acfa0e29c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a939b357abddb6d1e97222c24ee181e4df155b3224eb3e59855823990b9fc6a", "generation_seconds_batch": 21.908270057057962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2500059604644775, "target_z": 2.2343749831771733, "ce": 6.09409761428833, "genuineness": 3.9173994064331055, "identity": 0.7315018177032471, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6b5c154f01c98126828b5ba8f592337cd1f03388fcb5d322255e029acfa0e29c.mp3" }, { "id": "heldout-es-3", "language": "es", "sentences": [ "Alguien ha movido la pequeña silla azul.", "Estaba aquí... junto a la estantería alta.", "La colocaremos después de comer." ], "text": "Alguien ha movido la pequeña silla azul. Estaba aquí... junto a la estantería alta. La colocaremos después de comer.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-3", "key": "6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f3997c24282573437b5f49a5aab8e614bf82fb409c69dfab1a71525d2f20cc0", "generation_seconds_batch": 16.792581256013364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 6.181467056274414, "genuineness": 4.369274139404297, "identity": 0.6321887969970703, "hypothesis": "Alguien ha movido la pequeña silla azul. Estaba aquí, junto a la estantería alta. La colocaremos después de comer.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "6bd7e699b5d3a38b7d51448ab33af017b1eefa0cf2bd90f8bdbf84a1eab32bc4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5739d3b332cefc0cd43e85e79085c31d8788ffa146359f80de27718b8456e13f", "generation_seconds_batch": 16.959154475014657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.8397622108459473, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6bd7e699b5d3a38b7d51448ab33af017b1eefa0cf2bd90f8bdbf84a1eab32bc4.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "6beac3e1fa73f2f6979344f10f64de927a31e5ccefd6a0ee735dd93144f6285a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "00ba559801467df7a7be24ffaacfbf012fa5476205cbd4136c20e50306c05c10", "generation_seconds_batch": 16.801777162007056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.419747352600098, "genuineness": 4.193545341491699, "identity": 0.7235175371170044, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6beac3e1fa73f2f6979344f10f64de927a31e5ccefd6a0ee735dd93144f6285a.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "6c00a37f7564d44b2e61681a93e9f1bdd176c281e93f21782294d66dd0e2ec9f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ff0c525e03f6937747cd473a988864e018972726595594cd9f71e1cb88711a3", "generation_seconds_batch": 16.43211724993307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.7922817468643188, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Dis-moi quand tu es prêt.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6c00a37f7564d44b2e61681a93e9f1bdd176c281e93f21782294d66dd0e2ec9f.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "6c1e29f36da53c7557aeb9d6c7a9adee726bf9c5a00adbad41d246fe18ffce59", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b243242a4b80dee83b02847ed3115b6f65342bc634843417f11206850576dd14", "generation_seconds_batch": 17.33204422495328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.268836498260498, "genuineness": 4.6203155517578125, "identity": 0.7818688750267029, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6c1e29f36da53c7557aeb9d6c7a9adee726bf9c5a00adbad41d246fe18ffce59.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "6ccf28ea098f0b8e11204bea19833972d8cf56b2e591cb9e05ddc9d525e3f77e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "233b3e5ee9ebde3cc8e83e2f2b5f05b9a690d58a24cf6e17b3b0ef1ace34952a", "generation_seconds_batch": 16.053499951958656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.165581703186035, "genuineness": 4.394378662109375, "identity": 0.7252591252326965, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder, tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6ccf28ea098f0b8e11204bea19833972d8cf56b2e591cb9e05ddc9d525e3f77e.mp3" }, { "id": "heldout-es-4", "language": "es", "sentences": [ "Escribí el número en el sobre.", "Léelo bien... y comprueba las dos últimas cifras.", "Llámame cuando tengas un momento." ], "text": "Escribí el número en el sobre. Léelo bien... y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-4", "key": "6ce4b9a5397f9896897cd45dcb048f5a6c86ed9bb56ac93f60d51f9184fa53d3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Escribí el número en el sobre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e7dac6146556f1cd72036ee4ad83577257e0b316f1ec71a5ce477d3764f6ea3", "generation_seconds_batch": 18.84678244800307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.284721851348877, "genuineness": 4.068024158477783, "identity": 0.6493072509765625, "hypothesis": "Escribí el número en el sobre. Lélelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6ce4b9a5397f9896897cd45dcb048f5a6c86ed9bb56ac93f60d51f9184fa53d3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "6d0e2e0a7c80f024a447afdd636b0102f2b615d731138c2a13408c5779d62608", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2ccb3d2ce34c7058857fcbf17e6b2c0a95c1db13ddf655978a2521b3cbba764", "generation_seconds_batch": 16.809752718079835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.23706579208374, "genuineness": 4.3881025314331055, "identity": 0.6451541185379028, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6d0e2e0a7c80f024a447afdd636b0102f2b615d731138c2a13408c5779d62608.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "6d0f2def2ae596e711bd33fc20a2fa637fb78fdd108a6a0ce530e8cf4216e56e", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fed6fb19d5cad51c4dd8d892471d48e1834a2631fa6e9adc14e9dd5bb3c4d61", "generation_seconds_batch": 19.403771906974725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.308549880981445, "genuineness": 4.381826400756836, "identity": 0.4082222878932953, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6d0f2def2ae596e711bd33fc20a2fa637fb78fdd108a6a0ce530e8cf4216e56e.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "6d2f0c359c0965365a89b0e4db36bf1354a759174a3544af9c50f3970862c158", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b47fe7a4ca17500a2efd2f2c366e30f7aca472cc8dffbeaf166bf2bf69379c4", "generation_seconds_batch": 17.215266718994826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.2767791748046875, "genuineness": 4.369274139404297, "identity": 0.6873778700828552, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6d2f0c359c0965365a89b0e4db36bf1354a759174a3544af9c50f3970862c158.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "6d818e270846737388d90877eec0f44948a2863dddf2239e734162f3090cd2d3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb5dcf86de8a327b5e7187da699fee44cdb27ab6be23d9c41996c4276a3e138b", "generation_seconds_batch": 24.472987315966748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.2767791748046875, "genuineness": 4.2249250411987305, "identity": 0.37291526794433594, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6d818e270846737388d90877eec0f44948a2863dddf2239e734162f3090cd2d3.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "6dc0efae3b76f059ec54570dc12d277658977220199a4d25076c8523f5d05aaf", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a47db0e894d3f42b5cf26608d88598167d7a9f96ff9b7d5f049cfff52f88312", "generation_seconds_batch": 20.795082685071975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.467403411865234, "genuineness": 4.281409740447998, "identity": 0.3572959899902344, "hypothesis": "Die Straße führt hinter der Brücke nach links, geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6dc0efae3b76f059ec54570dc12d277658977220199a4d25076c8523f5d05aaf.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "6e37b6318544843a7cf3a47e909860177a5c730d5b72c373c3ad2f7a462127e9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b4f31b54f4d345fff7d73567a6203b58b84f3bcba58ae60280c14a762711f7f", "generation_seconds_batch": 23.855075791012496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9550375938415527, "target_z": 1.9921875527625015, "ce": 6.252951145172119, "genuineness": 4.2751336097717285, "identity": 0.6343188881874084, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6e37b6318544843a7cf3a47e909860177a5c730d5b72c373c3ad2f7a462127e9.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "6e3d2ff52e1548472a5504b788bb604ff735da4bd1e25dfda5eef3459201961f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9d0ed702b71a553b92b41a09adb0a87ceae64011713d31fe4dd87ebceac87b4", "generation_seconds_batch": 14.638222874025814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.459460258483887, "genuineness": 4.306513786315918, "identity": 0.5987088680267334, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6e3d2ff52e1548472a5504b788bb604ff735da4bd1e25dfda5eef3459201961f.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "6e4b4dfc2e4dc6ff164688f217e09b1833cf1dee0114ce5b73f7f5d425cc3675", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a5f065894a315a5755093c5963c3a2deac59ed28f598d352096e96be37a2a5e", "generation_seconds_batch": 23.173846847959794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.1894097328186035, "genuineness": 3.8420867919921875, "identity": 0.3140552043914795, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6e4b4dfc2e4dc6ff164688f217e09b1833cf1dee0114ce5b73f7f5d425cc3675.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "6ea389844ed1dd7d263010e41e4b8685c043ce6977d0974430b716be232035dd", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5db03f4638de79c9dea6d14c9693bded38fa97915c61561109ab7bda367b6741", "generation_seconds_batch": 15.541045977035537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164291858673096, "target_z": 2.4531250718793496, "ce": 6.157639026641846, "genuineness": 4.030367851257324, "identity": 0.47007259726524353, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6ea389844ed1dd7d263010e41e4b8685c043ce6977d0974430b716be232035dd.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "6ec929da3b7dcf0f528a9e4e2f14346d39cb43ebc30d7e81e220786482143c8d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d56180b9a092135e6ef9d91b91137a27a274b48b7b662051e177bd4411b7a80", "generation_seconds_batch": 19.25936669902876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 5.959072113037109, "genuineness": 4.193545341491699, "identity": 0.6070877313613892, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6ec929da3b7dcf0f528a9e4e2f14346d39cb43ebc30d7e81e220786482143c8d.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "6f81a5d908c584973da51ee22172f6519150614da54d90ac798769687a48edc7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b29950f899f7d0419ee4980de442548823a83e939b9e00b059f6cf091d0007d", "generation_seconds_batch": 26.03256964602042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.24500846862793, "genuineness": 3.9425034523010254, "identity": 0.37275761365890503, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/6f81a5d908c584973da51ee22172f6519150614da54d90ac798769687a48edc7.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "7020954bac55aa070be8b5911a06609263ea93725092f1fed755c0f25d84ebb8", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0a1716395e1c3135ad06dbc68b32d7a08b9ffb18df56831a5ffdadaf485a3bb", "generation_seconds_batch": 17.790041718049906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.229123115539551, "genuineness": 4.1998209953308105, "identity": 0.8317288756370544, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7020954bac55aa070be8b5911a06609263ea93725092f1fed755c0f25d84ebb8.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "7030abfa69d668712efec68561b2260ccb8ea70fdecd970c12676c1f5ac90f15", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "569cb0c2c64be98395cc61f4a07d9ff25e241f43b8b2684ee6567bb6ee664ccc", "generation_seconds_batch": 16.847032010904513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.252951145172119, "genuineness": 4.193545341491699, "identity": 0.6399742960929871, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7030abfa69d668712efec68561b2260ccb8ea70fdecd970c12676c1f5ac90f15.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "7041388c032ad6ebdbe175bf82bb70e6fc50a9ab4ccc6524993f42c648c9aa98", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8e490553f73b764d0b03f8d14b83c4db4b2d1f408d69340ed0ad9c3a7ea981e", "generation_seconds_batch": 23.090065954020247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3555855751037598, "target_z": 1.5000000489391319, "ce": 6.213237762451172, "genuineness": 4.218649387359619, "identity": 0.7831077575683594, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7041388c032ad6ebdbe175bf82bb70e6fc50a9ab4ccc6524993f42c648c9aa98.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "70a3edaacab16cf1ca392c7d4baea0fe61f70aaff3c1d0c4022c7a7d2c248df7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3128cd5673bff502f84bd94f8289ef9a248f93787a13cc811f25e6019e2736d4", "generation_seconds_batch": 22.997103448957205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.24500846862793, "genuineness": 3.8169827461242676, "identity": 0.1625966578722, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/70a3edaacab16cf1ca392c7d4baea0fe61f70aaff3c1d0c4022c7a7d2c248df7.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "70a94a5dc40437e6735db7e08f2ee54ccd9931d5fcee4082b58d976d4e7a0a52", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d353fa70da7f811dd6b0e9dc6a37240cdab28a25fc3bb100a0cc01d8040d7590", "generation_seconds_batch": 25.358862838009372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.4693537652492523, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/70a94a5dc40437e6735db7e08f2ee54ccd9931d5fcee4082b58d976d4e7a0a52.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "70cab7eab57e66e6434dc8ee293b526e04c60a60133768e5807bdae445e26579", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44823febf6c00d2c11217705d11217ba14041ae996036665e2e3c23c91f36232", "generation_seconds_batch": 15.902370056021027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 5.967014789581299, "genuineness": 4.243753433227539, "identity": 0.5201050639152527, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/70cab7eab57e66e6434dc8ee293b526e04c60a60133768e5807bdae445e26579.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "7136c1a6216181db352de250fda4b1ca8290c24f2aad3bcb847ecfae427c7378", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddb8cd2bbde3e112a6ce424deec2f4cd9f0582a38150ca7d1f2f7e25ec5f1292", "generation_seconds_batch": 14.933941408060491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0415868759155273, "target_z": 1.2421875282929355, "ce": 6.308549880981445, "genuineness": 4.331617832183838, "identity": 0.6009376049041748, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7136c1a6216181db352de250fda4b1ca8290c24f2aad3bcb847ecfae427c7378.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "713d3ab3a1118a65cf949cf9d0e78ae50e39833b0a928509010ca1260caca1af", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eaa5857926050fca21cb0730b2e6ac1aaf997e312317a58f60b0217f61497f29", "generation_seconds_batch": 16.889136526035145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.284721851348877, "genuineness": 4.149612903594971, "identity": 0.49057310819625854, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/713d3ab3a1118a65cf949cf9d0e78ae50e39833b0a928509010ca1260caca1af.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "7172e2db19145cde06087af1f1adc3e40f8e1fc12f24e02ec1ef13081e3dfe16", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "295bd6688c313998225d676f6018d6b0008cbabf6a3790d3cfee8c31f9d1b443", "generation_seconds_batch": 21.500453840009868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.28675955533981323, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7172e2db19145cde06087af1f1adc3e40f8e1fc12f24e02ec1ef13081e3dfe16.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "71738982bf824215cedd23dfd6abf39c959c55226056d688ae8b27a8182eff38", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f641f0589ba1be2d5d74c1aba33bc3dd7249fa47d8d8e828518d38a1b64b5a7", "generation_seconds_batch": 25.761639722040854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9178904294967651, "target_z": 1.1406250290576094, "ce": 6.435632705688477, "genuineness": 4.356721878051758, "identity": 0.6014077663421631, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/71738982bf824215cedd23dfd6abf39c959c55226056d688ae8b27a8182eff38.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "71ce61ce51a7761cc344fc30954232782595895d8da03a500b1ed783e5f82b73", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8bd13144e11276c880721b44c43f2eae310178f85f22079f09d8de57fe32ce4", "generation_seconds_batch": 25.922847214038484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.133810997009277, "genuineness": 4.607763767242432, "identity": 0.5183297395706177, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/71ce61ce51a7761cc344fc30954232782595895d8da03a500b1ed783e5f82b73.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "7242b82a5b68a7fd44b4c21ff530937541841d1265d722b7213c0dcdd4ada3ad", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6890807891a8b95f586eee91f662eb6e3e52def15c5026dcfa788c22318648c5", "generation_seconds_batch": 23.234571039909497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.284721851348877, "genuineness": 3.998987913131714, "identity": 0.3030470609664917, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7242b82a5b68a7fd44b4c21ff530937541841d1265d722b7213c0dcdd4ada3ad.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "729659a8491b365bda197dbd52181272ccba1ded1d10c7ec333fb09596e50663", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2402cf26705b403fc932312d1410c8bcad1750e48f8bce9a4b757dd714a7c69e", "generation_seconds_batch": 22.774087387952022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.149696350097656, "genuineness": 3.8922953605651855, "identity": 0.3956795930862427, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/729659a8491b365bda197dbd52181272ccba1ded1d10c7ec333fb09596e50663.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "72c15e49141cb96783de7f575e8f8894d0dfa8e8678704b1ab427082ac8edb55", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2316390cf077dd622b63d8fe36b437ddcbfcc2453573c2862d48bc3af5d6e746", "generation_seconds_batch": 17.917920239036903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070132255554199, "target_z": 1.2656250657619583, "ce": 5.95112943649292, "genuineness": 4.168440818786621, "identity": 0.14058446884155273, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/72c15e49141cb96783de7f575e8f8894d0dfa8e8678704b1ab427082ac8edb55.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "72c8dd045ef81be090102a6d6db50fec22a91c687370c82139530ad2cb2637d9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e63a2316b0bd35e8bd8d6e3b26e5183fae764f37e3c3d278a8eb183d16e78274", "generation_seconds_batch": 22.580291702994145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 6.141753673553467, "genuineness": 4.582659721374512, "identity": 0.6810446977615356, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/72c8dd045ef81be090102a6d6db50fec22a91c687370c82139530ad2cb2637d9.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "72e96da4b1eb385228000e7169e95cdbc178e723cdac5f851771a18e7b41af66", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6964d5ac2ac0e7d5240f5c14d012757ed32c80531a5f3ff709b1872c2a40b527", "generation_seconds_batch": 23.546610806020908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.205295085906982, "genuineness": 3.7416701316833496, "identity": 0.4291146695613861, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/72e96da4b1eb385228000e7169e95cdbc178e723cdac5f851771a18e7b41af66.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7396b4093302d6e8230b6af47d9a60bfb78afeb1c4b5d317c3be53315158c0f4", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4965d0ee6f492dafac87d73f17d9aa475f843723e562202effbc71f8c9f4f305", "generation_seconds_batch": 17.366462100995705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2518326044082642, "target_z": 0.5937500030586957, "ce": 6.348263263702393, "genuineness": 4.444586753845215, "identity": 0.37250375747680664, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7396b4093302d6e8230b6af47d9a60bfb78afeb1c4b5d317c3be53315158c0f4.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "73ae6833271b43eb944cfebfb19b12dc9b555b93c9bdd246ba16f59dd52d8577", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d1090ecae59872bcaaef2efe79df96000c592b54edac5bde2abc8f22940950a", "generation_seconds_batch": 21.637516026035883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.459460258483887, "genuineness": 4.482243061065674, "identity": 0.6009594202041626, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/73ae6833271b43eb944cfebfb19b12dc9b555b93c9bdd246ba16f59dd52d8577.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "73e3ca56372fe9a28cef5f9753fafcb2fbd72aecce271c3e51aed437d33f8cd8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8efd5d24228c9a9e8424bb70200505c25ebc636ca78c698c7f2c063297281155", "generation_seconds_batch": 21.734900022041984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 5.895530700683594, "genuineness": 4.074300289154053, "identity": 0.5502022504806519, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/73e3ca56372fe9a28cef5f9753fafcb2fbd72aecce271c3e51aed437d33f8cd8.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "742bfe5f579717664473e79671793a3863631d4ec63487c17c9aede0de039f3c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e577b55cdf074cf1b1cdcde922b39d8ea6fc8575f387c68bc70aaf8768e3bfe6", "generation_seconds_batch": 25.761639722040854, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460251808166504, "target_z": 1.5859375558211972, "ce": 6.324435234069824, "genuineness": 4.268857479095459, "identity": 0.5647991895675659, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/742bfe5f579717664473e79671793a3863631d4ec63487c17c9aede0de039f3c.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "743a62e38caf82163652a2049b7471cfe8459699ae599e2040b9550564e6dd7c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5605f2f76abaabd2be2a029fe6fbded0f96a9351587ea3b359ce31e31f782557", "generation_seconds_batch": 21.340411863988265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.1894097328186035, "genuineness": 4.137060642242432, "identity": 0.5524769425392151, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/743a62e38caf82163652a2049b7471cfe8459699ae599e2040b9550564e6dd7c.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "74434e5f0da617f01db22ecffca3a3c004c8ece457235ee9b2f28cbeee0626f9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70725b05d64ccca75dcd9a420051ee0f607f0ec5aeb7d3321fa50947af9d0464", "generation_seconds_batch": 22.725412467028946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.475345611572266, "genuineness": 4.143336772918701, "identity": 0.3031094968318939, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/74434e5f0da617f01db22ecffca3a3c004c8ece457235ee9b2f28cbeee0626f9.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7475869ce442c074cb4b4a0092cddf7bcd21af4951e16261cfcb5e929f094588", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d10719806d0be6a67b4eb3a37eee38f975ae3948c3b26bc29495a10a6b5e093", "generation_seconds_batch": 17.042026398004964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.794193983078003, "target_z": 1.0390625298222833, "ce": 6.324435234069824, "genuineness": 4.381826400756836, "identity": 0.49668556451797485, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7475869ce442c074cb4b4a0092cddf7bcd21af4951e16261cfcb5e929f094588.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "747c0a1690588e2e66bfb195f5fcad2d114beb8a5e22636d2f50bdafe1a67f51", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9c65a7fa6c2af10a39cbc2f0b3a6c180bef03cad759aa48521d1a683cbae9075", "generation_seconds_batch": 20.11246775498148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.316492557525635, "genuineness": 4.218649387359619, "identity": 0.42105185985565186, "hypothesis": "Johnson fandnich, die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/747c0a1690588e2e66bfb195f5fcad2d114beb8a5e22636d2f50bdafe1a67f51.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "7498c773df6b50af796e860c18c9ceb0b370709a6f43e9f7f1a4c5123659df6d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebfa7324127905dfe260fb8a5f45b14b0af90f9e3e0bc2acfe627bec011313a7", "generation_seconds_batch": 18.911488098907284, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.098677635192871, "target_z": 1.289062603230981, "ce": 5.998785495758057, "genuineness": 4.068024158477783, "identity": 0.667726993560791, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7498c773df6b50af796e860c18c9ceb0b370709a6f43e9f7f1a4c5123659df6d.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "74e8a4270da1f2a98dbadb8e5faaf0b23d3974629939d1b210ac5f97ae7b2f27", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47e4c4532b0a2588e4f37d8f7975d3e53da4cb22ad0a9b616cd8185878f50608", "generation_seconds_batch": 22.365785385016352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.38003396987915, "genuineness": 4.419482707977295, "identity": 0.4711243510246277, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/74e8a4270da1f2a98dbadb8e5faaf0b23d3974629939d1b210ac5f97ae7b2f27.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "751e94644c0a392efe64e76fb0c38e07242ed58a2bddab4c9722e098de99e7cf", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a6ef43ce003ea985f29bddbb7a11f6b3a9cacfb0a952eaf0d6e3dfb1975a7b9", "generation_seconds_batch": 17.33814301295206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.0146708488464355, "genuineness": 4.124508857727051, "identity": 0.8067507743835449, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/751e94644c0a392efe64e76fb0c38e07242ed58a2bddab4c9722e098de99e7cf.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "7567a73750afd9a5540344a5f62defe57c742bcaafc5c631ed2a98bc389e1ac4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d544f7caa4a3ad72aa8e4f18b93cf04c85d84b1125d628840a516836aa29ef15", "generation_seconds_batch": 26.58764007501304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.38003396987915, "genuineness": 4.419482707977295, "identity": 0.44439977407455444, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7567a73750afd9a5540344a5f62defe57c742bcaafc5c631ed2a98bc389e1ac4.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "75d10af7f1f56fb30f7bd6f1515f5945393c152783aa7ad59a4e8e038c92df8c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "361780a86e01d06ef15a271a695b68f3a786227371b005e2d08a2874a4da999d", "generation_seconds_batch": 28.033079813933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.268836498260498, "genuineness": 3.9299514293670654, "identity": 0.2536928653717041, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/75d10af7f1f56fb30f7bd6f1515f5945393c152783aa7ad59a4e8e038c92df8c.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "762f17fc1e63f2e82e7441aa27e1482e5cf6375c7aede1a5c14ed8904eb1e8e4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66074ae4f5764c4000a58d0ee7855c4bea310e8c60587a5e4a02f76f2ee1911b", "generation_seconds_batch": 22.951103447936475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.09409761428833, "genuineness": 4.068024158477783, "identity": 0.443495512008667, "hypothesis": "Gießt, Ban von Dis, Ban Pirgin von Sat. Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.42105263157894735, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/762f17fc1e63f2e82e7441aa27e1482e5cf6375c7aede1a5c14ed8904eb1e8e4.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "771ad73f82699697eaadd3432942c8d730e760d369f1028919ac4fadeedf83ce", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9b7dfa76f6345f745ad860e798a4e5ff4cd9bcec318a68d22b5244a97da0afe", "generation_seconds_batch": 20.057066308916546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.260893821716309, "genuineness": 3.9299514293670654, "identity": 0.6070920825004578, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/771ad73f82699697eaadd3432942c8d730e760d369f1028919ac4fadeedf83ce.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "775c133ae7c1f7e5ed2119244ddf187f73ca9291c4d9e30f1b93b960fbc4c61c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54d8d891053aaf11f5d82104c2a74116aed5b34b44b3cf2047b1ac6b47e0bcf7", "generation_seconds_batch": 23.791357461945154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 5.099277973175049, "genuineness": 4.074300289154053, "identity": 0.5836126208305359, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/775c133ae7c1f7e5ed2119244ddf187f73ca9291c4d9e30f1b93b960fbc4c61c.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "7761a6f37dd3843780dcdba4b234af164507df22edc19b59c4c0e78ed4e5b50c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5527e2a96189d67d5d672bb1080e0e6b2bc6b184260960ad9b3582eebd9a166", "generation_seconds_batch": 21.94106747605838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.1894097328186035, "genuineness": 4.030367851257324, "identity": 0.2372710406780243, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7761a6f37dd3843780dcdba4b234af164507df22edc19b59c4c0e78ed4e5b50c.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa4bbb8a60e3ed891f8582d74ec2e6d4a616cbab5c9695025a88e9fc964c728d", "generation_seconds_batch": 16.95426589192357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.6189730763435364, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "77f24f979503750fb6bd33b991e5187e6d8feab485704d5f6dced7622e152be4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c03452f3052b562c087c7a33d8f4b01ff6cdbfd5c67aef1ecdc128c7cb661788", "generation_seconds_batch": 22.958582942024805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.332377910614014, "genuineness": 4.2374773025512695, "identity": 0.4561415910720825, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/77f24f979503750fb6bd33b991e5187e6d8feab485704d5f6dced7622e152be4.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "77f5f34074ae1aa4ab6aa30564512d7a0c78c0eeff1e1f6897846b4fc51c1ca6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0ab73f5ff98bb6788e72cffdf1faab24a8e5163c62a62f0cf3ff49707563300", "generation_seconds_batch": 16.12785638589412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.561073660850525, "target_z": 0.8476562022078792, "ce": 5.903473377227783, "genuineness": 4.532451152801514, "identity": 0.7059003710746765, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/77f5f34074ae1aa4ab6aa30564512d7a0c78c0eeff1e1f6897846b4fc51c1ca6.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "780952cb16c7cc428fefe0542ea79487386762609653f5db8286f02b0d034f75", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adb01a2d95f64c06bbeb152020c9b8a3f228ed391851c8464000b03cec4d6bce", "generation_seconds_batch": 16.72048512008041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.213772177696228, "target_z": 0.5625000183521744, "ce": 6.260893821716309, "genuineness": 4.65797233581543, "identity": 0.5414443016052246, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/780952cb16c7cc428fefe0542ea79487386762609653f5db8286f02b0d034f75.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "786e6fa5d1cf49e12b5325410457a5b9fa2514eaf3bc0f9108c80d64a91212f3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f02ae5522a5c4b50f0511ff3f2f319f80e8a76d1f0b06f742788d49da6f59f3", "generation_seconds_batch": 22.39556314703077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.046441555023193, "genuineness": 3.9048471450805664, "identity": 0.5556468963623047, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/786e6fa5d1cf49e12b5325410457a5b9fa2514eaf3bc0f9108c80d64a91212f3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "789f0c477c08da79dcca34eadf417ec79dea239f249bec07962f46265a921c34", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c95363cb2c66688d101ed358bee946e149c6a6b6d49ee2c35d08f42ddaa39629", "generation_seconds_batch": 22.195605981047265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.1020402908325195, "genuineness": 4.18099308013916, "identity": 0.589134693145752, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/789f0c477c08da79dcca34eadf417ec79dea239f249bec07962f46265a921c34.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "78ca3cc871c632ace7e8c48a186f6c430a4e2266d89348c3d1ca610fe260cf34", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3ce35c4a3c5219726b9e3f3cfc9417286e10264740f319926b65e5330b9d48a", "generation_seconds_batch": 23.23506689397618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.086154937744141, "genuineness": 4.118232727050781, "identity": 0.734624445438385, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent. Dat shall.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/78ca3cc871c632ace7e8c48a186f6c430a4e2266d89348c3d1ca610fe260cf34.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "78fc14d720e4a09886f1edfa10719023208c78f4c3359e77acdf948aae5e964c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a1c9eea96bc43a90f23411dd1c0e8c6daeddb768491553c5a41d342a38c2b17", "generation_seconds_batch": 17.481961090001278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.046441555023193, "genuineness": 3.9550557136535645, "identity": 0.706753134727478, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/78fc14d720e4a09886f1edfa10719023208c78f4c3359e77acdf948aae5e964c.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d2ee54c2192eec3a44d59edf4d197d35bb3e3e283a5e521a6c3eebc2df07391", "generation_seconds_batch": 16.707111887983046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.630610466003418, "target_z": 2.5468750259989137, "ce": 6.1020402908325195, "genuineness": 3.998987913131714, "identity": 0.6980760097503662, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "79b7a29d82f051e6238c10328311827de0557b9eacbdb660180152a70852a518", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3209d02dee68d7ae39efa1ef08ddb3f1376d8bdd30f515f1013f1ddb7e944ff7", "generation_seconds_batch": 22.951103447936475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 5.967014789581299, "genuineness": 3.6789097785949707, "identity": 0.5163843035697937, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/79b7a29d82f051e6238c10328311827de0557b9eacbdb660180152a70852a518.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "79c37961370c0d27b137656a4e2207d91c081116f8af0cf730d153af4f59e84b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60dddda4d79ff33686af0d4dbe021c825f9b5d0022903df5e3addeec16b3c3af", "generation_seconds_batch": 21.112544320058078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.213237762451172, "genuineness": 4.18099308013916, "identity": 0.7282997369766235, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/79c37961370c0d27b137656a4e2207d91c081116f8af0cf730d153af4f59e84b.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "79e60105d7fd33548605fcbdd933e397fc02f0c1fb41392e0afceaeb94966f0a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1211a33ba1cc1088a62a072da38a8e0d9e080408645e753fde71b4cac12c311", "generation_seconds_batch": 22.42488304001745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.157639026641846, "genuineness": 4.13078498840332, "identity": 0.5806416273117065, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/79e60105d7fd33548605fcbdd933e397fc02f0c1fb41392e0afceaeb94966f0a.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "7a0abc9e47d7cc7be0587ec1e2ecef70a8d22bbc77dacdecf60cb0856875c730", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0706a4cdd97530a888278813a7d75f00a862f8fb16d1479ec60a3f01eef8e945", "generation_seconds_batch": 25.36704861500766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.205295085906982, "genuineness": 4.093128204345703, "identity": 0.4019470810890198, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7a0abc9e47d7cc7be0587ec1e2ecef70a8d22bbc77dacdecf60cb0856875c730.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "7a273ab4403180002e016c65602cf92652b61be67079ccf1aa15426329d3cb6c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c4680dc0e45cc2fdfef3d7674bbfba39e8f38588af02674eb084d2dcf7f4323", "generation_seconds_batch": 23.023433946073055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.173524379730225, "genuineness": 3.7667744159698486, "identity": 0.31937965750694275, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7a273ab4403180002e016c65602cf92652b61be67079ccf1aa15426329d3cb6c.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "7a45204eb89fcf73a6d8809d2aaa1c2c35cd6523d9071e2a6c62a3a3eb553858", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18938a1cc4d5afa8791170d7540fe7ba886fe9d57506c31c1593c82aaf96eedf", "generation_seconds_batch": 20.88879332097713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.38797664642334, "genuineness": 4.093128204345703, "identity": 0.4620639681816101, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7a45204eb89fcf73a6d8809d2aaa1c2c35cd6523d9071e2a6c62a3a3eb553858.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "7a7d8dda9075653c9fa72fe2522fef19f6f145cf72a931eed11597cd0acad100", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "849ad2f7794872f0195727a801b10f247d2b84140b6a4e244c15b68ff473acf6", "generation_seconds_batch": 22.10417850909289, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070132255554199, "target_z": 1.2656250657619583, "ce": 6.427689552307129, "genuineness": 4.494794845581055, "identity": 0.683745801448822, "hypothesis": "Ich habe die kleine Schacht gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7a7d8dda9075653c9fa72fe2522fef19f6f145cf72a931eed11597cd0acad100.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "7aa1c3398811e147078972269b8e1741778509c985179553d017fed1d14a87a3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e405a046f1a000bb71176c4b7c6875a612266a75a19cad67cc5ffaf86af9841", "generation_seconds_batch": 28.033079813933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 6.41180419921875, "genuineness": 3.798154592514038, "identity": 0.030882487073540688, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7aa1c3398811e147078972269b8e1741778509c985179553d017fed1d14a87a3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "7ac98155b2254577e9f619d485e62941cdb3190caa0b5255b8841473a9104773", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d35bccd7e27c6afcb5461e7f569b247b3612486647457089d43d492d3680e92", "generation_seconds_batch": 21.535951316007413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2500059604644775, "target_z": 2.2343749831771733, "ce": 6.1894097328186035, "genuineness": 4.2751336097717285, "identity": 0.5519388914108276, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7ac98155b2254577e9f619d485e62941cdb3190caa0b5255b8841473a9104773.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "7b12360748895aa45bff74d9fb06aaed6c39a82229d088be539fb6daeeb51f21", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41f79cde809f28ee56f9e50f7a9a5bf17aa4bf8de47a327936c0a796f99c65ab", "generation_seconds_batch": 23.383737852913328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 5.935244083404541, "genuineness": 4.086852550506592, "identity": 0.8181213140487671, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7b12360748895aa45bff74d9fb06aaed6c39a82229d088be539fb6daeeb51f21.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7b22e279d457cb46e43158f4859818a940ed165cb083068298260d6972d30282", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22449af6bad5183fbc83538a3ceba23f829b349d62258285854b032571f1a456", "generation_seconds_batch": 21.933430413948372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.205295085906982, "genuineness": 4.293961524963379, "identity": 0.6016899347305298, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7b22e279d457cb46e43158f4859818a940ed165cb083068298260d6972d30282.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7b260db7581e6103951c0d883a4384147b369546a4e33367b9e4762954911635", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ac9929a6a7f3431a4fe40d38ab74b367f00de5a5ed18a1672def49eae173421", "generation_seconds_batch": 23.735393090057187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9464356899261475, "target_z": 1.1640624686483687, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.2942902147769928, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7b260db7581e6103951c0d883a4384147b369546a4e33367b9e4762954911635.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "7b3dda7b67c1b572d218511533ed9c9f6706bd27dedccb39dc10dead0bbd9510", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c8ac24c910f0dcb5424761324bd2ba26827e16c3df0ed25d035644fb77863875", "generation_seconds_batch": 20.26669136201963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7076447010040283, "target_z": 1.7890625542918492, "ce": 6.1894097328186035, "genuineness": 4.319066047668457, "identity": 0.8214765787124634, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7b3dda7b67c1b572d218511533ed9c9f6706bd27dedccb39dc10dead0bbd9510.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "7b6e1fff8cc6bbae6150940171023051c94e4325cddb89d754fc22791ba3b366", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ef81ae544275379fcd7da6abaa6232ed0395abc58a1d8e83531a939b474c0ca", "generation_seconds_batch": 22.958582942024805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.467403411865234, "genuineness": 4.20609712600708, "identity": 0.7069969177246094, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7b6e1fff8cc6bbae6150940171023051c94e4325cddb89d754fc22791ba3b366.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7bce52ead2dd22286d05d3690a4998188d10966c90872aeb5f818454c7ececed", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6287e7ef7911928ca3ac395a22b27264826ad77d0ab5c975bfaeba2c5ab6f9f3", "generation_seconds_batch": 19.223884980892763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.4676688611507416, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7bce52ead2dd22286d05d3690a4998188d10966c90872aeb5f818454c7ececed.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "7c7b14435e42adfe081eaac63c25c09a8eed9195a31c67ed70e9cdd436e77f04", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9a2ad738b4d00b13fa41b2b09a07941633209babc04b7daa2033422e4bc02ac", "generation_seconds_batch": 16.66987349302508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 5.796247482299805, "genuineness": 4.344170093536377, "identity": 0.6250232458114624, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7c7b14435e42adfe081eaac63c25c09a8eed9195a31c67ed70e9cdd436e77f04.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "7c9edeb2b6d9e4ab3be57bebed198958edf421eea4955acf91bf82e582909004", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15cf5b4d154b4180b2037d58f3f462af128c0a92182c97f57393026b8dcd0f2f", "generation_seconds_batch": 24.513505595969036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.136737823486328, "target_z": 1.3203123921809754, "ce": 5.95112943649292, "genuineness": 4.300237655639648, "identity": 0.7364966869354248, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7c9edeb2b6d9e4ab3be57bebed198958edf421eea4955acf91bf82e582909004.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "7d957ae46d97587f8e6163b34f562f91c74d30294c4463adfda45729676f898b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9192c51f7929b648cf661365aacf8d971d540c52e90df6782e0d0aa4d7482f21", "generation_seconds_batch": 25.76499787310604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.252951145172119, "genuineness": 4.293961524963379, "identity": 0.475666344165802, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7d957ae46d97587f8e6163b34f562f91c74d30294c4463adfda45729676f898b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "7dc192cf536aa62f36d526f554d9ec4b78f99a91681df03fd9ed145406a0b1d4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea86739e17343327ed74fcb0c83d6fd362396e3f53c4aa00aaf612b7b51d5de8", "generation_seconds_batch": 20.26669136201963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6505539417266846, "target_z": 1.7421874793538037, "ce": 6.197352409362793, "genuineness": 4.419482707977295, "identity": 0.6549232006072998, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7dc192cf536aa62f36d526f554d9ec4b78f99a91681df03fd9ed145406a0b1d4.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "7e2e3d4201362c7fce6f1446d2cbd48292de8e2f41ecbc3da3207b3602d243c2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91394350108e273446feae2fae95f7b60d162060cb9f785c02fff96afcdc5050", "generation_seconds_batch": 21.305803386028856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 5.831989288330078, "genuineness": 3.9927117824554443, "identity": 0.8410483598709106, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7e2e3d4201362c7fce6f1446d2cbd48292de8e2f41ecbc3da3207b3602d243c2.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "7e40ed3650643b33d503497b75a4e533ccab4d219755a236018af1adf42e1d8b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3c020416ffa79d7668ac099bf4026ad0d83ea7721a2cdf067a801e15e112ef4", "generation_seconds_batch": 23.418603440979496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.366014003753662, "target_z": 0.6875000550565232, "ce": 4.946381568908691, "genuineness": 4.369274139404297, "identity": 0.6185612678527832, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7e40ed3650643b33d503497b75a4e533ccab4d219755a236018af1adf42e1d8b.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "7e7a259ddba8a35f11dd3cf8a9d5a84d1e7751ddfe746dd90b4f08a1166f0847", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dcd4dfd45146a57fada8c024e8b6d4ca3186b661cc1ff8a92cf258931f9022a", "generation_seconds_batch": 17.050725597073324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.141753673553467, "genuineness": 4.193545341491699, "identity": 0.7826672792434692, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7e7a259ddba8a35f11dd3cf8a9d5a84d1e7751ddfe746dd90b4f08a1166f0847.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "7e88500d5c1c911118c4c77ab78a930308f9050cec988bff3658e7fa1ef49efa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eff7d5a419e6faf7e03aef57f342b32b02fa18f73d3815a734743703bff1e2bc", "generation_seconds_batch": 22.774058792041615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 5.998785495758057, "genuineness": 4.582659721374512, "identity": 0.7448318004608154, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7e88500d5c1c911118c4c77ab78a930308f9050cec988bff3658e7fa1ef49efa.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "7f1676de038421f411691c73e0b70ab38eb57ee960cfbb94e55e13b0e9d1d37b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41b1ab1b17804a23c57f223e321091ceaee58e535c360d4640239ce5da51018d", "generation_seconds_batch": 17.901635734946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.231889009475708, "target_z": 1.3984374518255422, "ce": 6.1894097328186035, "genuineness": 4.6956281661987305, "identity": 0.7201787233352661, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f1676de038421f411691c73e0b70ab38eb57ee960cfbb94e55e13b0e9d1d37b.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "7f2afeaa642263644162ca93a285cf6c4aecff9f88ced74b6b7b5ed589c6e91e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2222a87de10ef7ebd79ca8182f7e71ad18f53abea5e55010807bda6345f0db6", "generation_seconds_batch": 23.675143119995482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.09409761428833, "genuineness": 4.068024158477783, "identity": 0.784986674785614, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f2afeaa642263644162ca93a285cf6c4aecff9f88ced74b6b7b5ed589c6e91e.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "7f2f00452b4d4e9723026f0a2de7d86e883ada37d9893b113ae2530f8b120633", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96ddb4fd6e46beb9c6b4ac3ff813f26765afc22c225eb6be1c884d264806de25", "generation_seconds_batch": 16.895594342029653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.252951145172119, "genuineness": 4.2374773025512695, "identity": 0.7021652460098267, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f2f00452b4d4e9723026f0a2de7d86e883ada37d9893b113ae2530f8b120633.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "7f41e45ca46d873b0d4cb74245b1cd37ebc65ebf583f97e8b13fd482bfc3aaca", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfef4dfe2bbe66d11bcb51bd459914d0308258f36403b707ef4056a69620f635", "generation_seconds_batch": 22.018885001074523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.149696350097656, "genuineness": 4.074300289154053, "identity": 0.639388918876648, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f41e45ca46d873b0d4cb74245b1cd37ebc65ebf583f97e8b13fd482bfc3aaca.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "7f5542a8d19cb4a6c4362b686c352d5caa4a6a618803ee0a3c65e32b67f64a38", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "123b130452901fe35713fe3e3e286a92cdfe1ae08fee0e58131734b05fde8d18", "generation_seconds_batch": 23.383737852913328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 5.998785495758057, "genuineness": 4.111956596374512, "identity": 0.6051552295684814, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f5542a8d19cb4a6c4362b686c352d5caa4a6a618803ee0a3c65e32b67f64a38.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The keys are on the kitchen table.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca3c650cb2c337ca32998283da3f92ca6fa331466af99057e7546a7ea85f2e27", "generation_seconds_batch": 16.912574254907668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.395919322967529, "genuineness": 4.20609712600708, "identity": 0.2188417613506317, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "7f8b1f9a93e8e6a5861965f7095d61edcbd3ce130ed79833e5243f4e606b5670", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb88c08e9d7d84679829da8e7f206591310e173765c08889f870827a1b36640d", "generation_seconds_batch": 23.12579206400551, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.5227519273757935, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7f8b1f9a93e8e6a5861965f7095d61edcbd3ce130ed79833e5243f4e606b5670.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "7fb45c6c6bf1d4fdd0af44792588013b5301102b81074b652b7a4adaee7865c9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5962539a1942199dcbc095f26650d1638e524c600cb07a92fc10d21dcadb1716", "generation_seconds_batch": 22.675465599983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9369206428527832, "target_z": 1.1562500214108702, "ce": 6.324435234069824, "genuineness": 4.482243061065674, "identity": 0.21705681085586548, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/7fb45c6c6bf1d4fdd0af44792588013b5301102b81074b652b7a4adaee7865c9.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "800626e13033ede5697fd0f302af7aff466ebbc14fe5eac46667e107575b83fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b7f2080b86d066719da5e47e9f861391596628a9c77f6ccbc8c00419dacb515e", "generation_seconds_batch": 23.255556654999964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.41180419921875, "genuineness": 3.8797430992126465, "identity": 0.09008314460515976, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/800626e13033ede5697fd0f302af7aff466ebbc14fe5eac46667e107575b83fa.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "801795a376d4187ef3a5f1cbab4c34b2525ecc31d7c10bbbd0db3642e099deea", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2be8b7ac7fe9e4c62490e31413407d4b23a979752a381e99385e0e6cf90fe84", "generation_seconds_batch": 20.430139241972938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.252951145172119, "genuineness": 3.8295347690582275, "identity": 0.6504578590393066, "hypothesis": "Las llaves están en la mesa de la cocina. Mire en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.043478260869565216, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/801795a376d4187ef3a5f1cbab4c34b2525ecc31d7c10bbbd0db3642e099deea.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "804a5c4e72ff167429848e8a25479747d00b90ddc42f1690b71f795c45df97ec", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79559a68fb1ebb0ce53d2a5deabc805f5753344663b199c9d987eeaf9fbdb976", "generation_seconds_batch": 16.741789147956297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.24500846862793, "genuineness": 4.243753433227539, "identity": 0.5398470759391785, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/804a5c4e72ff167429848e8a25479747d00b90ddc42f1690b71f795c45df97ec.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe00194dae1153ac09bd5686405c4653b566805f37e51c80f8e40a5208226bc9", "generation_seconds_batch": 17.498806288000196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.149696350097656, "genuineness": 4.281409740447998, "identity": 0.705580472946167, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "80dfaaf5748fddba2a63a66d2c423279255dfe712225a529631bcf1038fe214c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d5907373fd8daac088c613cc6b17aa5fd7715e1974caaadd9a523e6d46fb08d", "generation_seconds_batch": 23.74840944202151, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.141753673553467, "genuineness": 4.369274139404297, "identity": 0.8614403009414673, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/80dfaaf5748fddba2a63a66d2c423279255dfe712225a529631bcf1038fe214c.mp3" }, { "id": "heldout-es-0", "language": "es", "sentences": [ "Tu mensaje llegó después de que me fuera.", "Lo leí dos veces... y guardé el teléfono.", "Dime cuándo podemos hablar." ], "text": "Tu mensaje llegó después de que me fuera. Lo leí dos veces... y guardé el teléfono. Dime cuándo podemos hablar.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-0", "key": "8147126523302030d22db06b8074199a9e4ccbc840fc1edf5103f66dadb8cf71", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df700e4bad57c8833850239f11a8e849ba8129d8971f6f790641fed741b81cfc", "generation_seconds_batch": 20.057066308916546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.1020402908325195, "genuineness": 4.030367851257324, "identity": 0.7319432497024536, "hypothesis": "Tu mensaje llegó después de que me fuera. Lo leí dos veces y guardé el teléfono. Dime cuándo podemos hablar.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8147126523302030d22db06b8074199a9e4ccbc840fc1edf5103f66dadb8cf71.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "818bf9a77c0098ab93e9dd8d08f486f82df1f5df7f60f822c00374ab33e52790", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d597682d6df50e868261f359d6d6d00bd71f26dd098f89815ad558e1745531f6", "generation_seconds_batch": 15.237069964059629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.062326908111572, "genuineness": 4.074300289154053, "identity": 0.6644308567047119, "hypothesis": "L'ARI tourne à gauche après le pont, continue, jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.1, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/818bf9a77c0098ab93e9dd8d08f486f82df1f5df7f60f822c00374ab33e52790.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "818d563aa0af9e2562734a73764a7aee41ec6d8d5889bdbfef42bc0650ac5505", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad968b60b2aee1fbe03a7f3bbe15012f8486e2a112409ab72832781dabb3c5d3", "generation_seconds_batch": 21.936667003901675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.4515180587768555, "genuineness": 4.545003414154053, "identity": 0.25540024042129517, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/818d563aa0af9e2562734a73764a7aee41ec6d8d5889bdbfef42bc0650ac5505.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "828a91eed2e09fbb4fc37bc594165720d30d127ff7a5650cd9ac3f697f666a54", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2773455ac0249495301f2761f962e3bcf6be776ae84be487a7c90b9d596dba5e", "generation_seconds_batch": 16.86874343000818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8037090301513672, "target_z": 1.046874977059782, "ce": 6.221180438995361, "genuineness": 4.632867813110352, "identity": 0.6318937540054321, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/828a91eed2e09fbb4fc37bc594165720d30d127ff7a5650cd9ac3f697f666a54.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "82aa231f76137cb375581b6b14e569d424dd4eb3c2d62a8ab0b4730e5cb7c5b9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f62ae0b693f9e9986e1145d6f1d7349d1f181262525c85e33385b8f91ee97bb", "generation_seconds_batch": 15.74870394996833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.231889009475708, "target_z": 1.3984374518255422, "ce": 6.229123115539551, "genuineness": 4.444586753845215, "identity": 0.7955414652824402, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/82aa231f76137cb375581b6b14e569d424dd4eb3c2d62a8ab0b4730e5cb7c5b9.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "82ad822c87d01878337dafd3e31b8dabeefd35455ad01c77e5d7ff4aa5aa37cf", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b37ba1b840c34fd1564497c00db7ac0bd1d85a172d2132542434702cf92abc35", "generation_seconds_batch": 26.68256098695565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.062326908111572, "genuineness": 4.507347106933594, "identity": 0.7275586128234863, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/82ad822c87d01878337dafd3e31b8dabeefd35455ad01c77e5d7ff4aa5aa37cf.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "837803bfd56fab527c2800d548ee4a52fe8d1183817992ded875be88da3f3e7a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "022fe70b51935c5cac2850c2280fddfc88c926728f581947f5e161051f41c115", "generation_seconds_batch": 18.467165970010683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.135824680328369, "target_z": 2.1406250290576097, "ce": 6.0305562019348145, "genuineness": 3.810706615447998, "identity": 0.5147941708564758, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/837803bfd56fab527c2800d548ee4a52fe8d1183817992ded875be88da3f3e7a.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "83d05981fc430df4ded7d18088a81f9f3d29d690a54caf29febd910e2b79e5f7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b19c7320012c4b34b274e1f486b0c9a364955879fd56d5776d118aa8aa6ddda", "generation_seconds_batch": 22.774058792041615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 5.824047088623047, "genuineness": 4.174716949462891, "identity": 0.801895797252655, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient, Boiscluspété, chez Frobatourille, le Bielia.", "wer": 0.2631578947368421, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/83d05981fc430df4ded7d18088a81f9f3d29d690a54caf29febd910e2b79e5f7.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "83fcfcb94bf206697b0e860f730df23f9d16d779274c9018662f929bd997ccdd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "333a85f18f54a91087011328848ebf5a5e9c604297e903b12ee9dcb3d7624ad2", "generation_seconds_batch": 26.620400486979634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.133810997009277, "genuineness": 4.231201171875, "identity": 0.6253544092178345, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/83fcfcb94bf206697b0e860f730df23f9d16d779274c9018662f929bd997ccdd.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "849d8310583101d1f766e6d1fe48830a79162680c9fd83ff22c94b7b178e7f8c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1af34bbd851f3e14045bf1c4490f7af88ef9b530a2cdc36d3a1d5cb11edb7df", "generation_seconds_batch": 23.78746170201339, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.2767791748046875, "genuineness": 4.457139015197754, "identity": 0.42210766673088074, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/849d8310583101d1f766e6d1fe48830a79162680c9fd83ff22c94b7b178e7f8c.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "84bdc7a2c134a7137438293eec6def538959f53f2f5622bbaa0cbfe716b715a1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7d758bd27a600add22aba04bb92264a28a985810fa9c5808ba29fcf5c09ce75", "generation_seconds_batch": 24.301791940932162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 6.1020402908325195, "genuineness": 3.810706615447998, "identity": 0.7553455829620361, "hypothesis": "La porte est encore ouverte. Le son nous est. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.2222222222222222, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/84bdc7a2c134a7137438293eec6def538959f53f2f5622bbaa0cbfe716b715a1.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "8556f136fe7eac801cc9c907394ce8c29a925f22d8bed0b9ceeab209f112363a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9e6518411a7db0557f15eebe4b9e428c4f588d075668afe947111c6887fd7ec", "generation_seconds_batch": 25.949483425938524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.364187240600586, "target_z": 2.3281249372967374, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.5806183218955994, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8556f136fe7eac801cc9c907394ce8c29a925f22d8bed0b9ceeab209f112363a.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "857880d0b01bc104806d41aa2f417e2624bffcebae2090465b9a79a9a0fb9128", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "edeff30bccbaa345cb58f1465505b6cf031a6add8e625ab85ea21e6468472bcc", "generation_seconds_batch": 22.95822083693929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.571735143661499, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/857880d0b01bc104806d41aa2f417e2624bffcebae2090465b9a79a9a0fb9128.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "85bb9be3d40969c7b5e98c2a654889279234ae63574538779ac977d9a090cced", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4b6e4ffad3e3ac8b0a4147d0d6c1574a5a386aa3ab165a21c8d5b4068cd5608", "generation_seconds_batch": 26.486166815971956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9550375938415527, "target_z": 1.9921875527625015, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.4169524610042572, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/85bb9be3d40969c7b5e98c2a654889279234ae63574538779ac977d9a090cced.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "85d834a43f8731d6f5574e41b528800ba126d77a4af63b003d3cac00c5c2ff9f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eed34481dd9760df7972554dcf78d0665a456f0eac5bb034642327adb173cdf2", "generation_seconds_batch": 23.022105142008513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.135824680328369, "target_z": 2.1406250290576097, "ce": 5.95112943649292, "genuineness": 3.8797430992126465, "identity": 0.6763990521430969, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/85d834a43f8731d6f5574e41b528800ba126d77a4af63b003d3cac00c5c2ff9f.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "85fc3b73e45b0db3d4524e95532cf3020a2270f8961e2b4044464e0ec9734439", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07b2fecf053fbb295654180a912128d15e96faefc7cbe2ecbb691c1d29f4010e", "generation_seconds_batch": 17.898199724964797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.6289290189743042, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/85fc3b73e45b0db3d4524e95532cf3020a2270f8961e2b4044464e0ec9734439.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "863b32f4577d2b9828cd75202cc84b2464b61bd6d15918c83688f562ba78d894", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "649ff90d513fa4299f1971ab6384e07ddbffc97ed05c69c68c3dd4e5c3c3af86", "generation_seconds_batch": 24.45141274004709, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.6209533214569092, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/863b32f4577d2b9828cd75202cc84b2464b61bd6d15918c83688f562ba78d894.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "8722c97cd4d7315c47545657eec9fbb6d931441fdfac4ed0105a2237459819dd", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a6f563d5c2c7e745e2e0d2b4390223b3292b473453391a0401c61c94de7c4c6", "generation_seconds_batch": 16.50531110400334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.300607204437256, "genuineness": 4.193545341491699, "identity": 0.29066455364227295, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8722c97cd4d7315c47545657eec9fbb6d931441fdfac4ed0105a2237459819dd.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "8727970b17e2a9924c1fbd6b294d3a8f01addf10f14a0952e9b0a25ee7d50755", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d4961ca75f8e39d2375622bf79dbf35ce508f47a827f54d803a293934cd6181", "generation_seconds_batch": 24.513505595969036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.0146708488464355, "genuineness": 3.7416701316833496, "identity": 0.5037949085235596, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8727970b17e2a9924c1fbd6b294d3a8f01addf10f14a0952e9b0a25ee7d50755.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "8760bf7524f3a3e515ea2018afc58cacb38a6966480f8766cbaf2adf407739c7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4113f25225604ef9ab1207250c34f652210669735670bda90aa98c5b7d5526e7", "generation_seconds_batch": 22.459264909033664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.24500846862793, "genuineness": 4.20609712600708, "identity": 0.6272138357162476, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8760bf7524f3a3e515ea2018afc58cacb38a6966480f8766cbaf2adf407739c7.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "87ea07f01a48b399fae035b36f86aa6946478001c33f10dbb72587c61bada82c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "582a9448a1cacfe44801009b956dfdc9ce5aa71253791db7533782541c99d76f", "generation_seconds_batch": 21.181950255995616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 6.149696350097656, "genuineness": 4.13078498840332, "identity": 0.6197380423545837, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/87ea07f01a48b399fae035b36f86aa6946478001c33f10dbb72587c61bada82c.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "881ccba20e1ce8c517a7356678dfbca9345b7323d1b9bff3d9993d903fb8d2e6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62f7c2cf7527ddcb59df83264f79e0b458bb5fde7b6cd46071ad6ffc0f19da87", "generation_seconds_batch": 21.637400410021655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886143684387207, "target_z": 1.7734374640603252, "ce": 6.205295085906982, "genuineness": 4.300237655639648, "identity": 0.6713625192642212, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/881ccba20e1ce8c517a7356678dfbca9345b7323d1b9bff3d9993d903fb8d2e6.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "88b57c6ceef1c14d00ab636f61f0649cdbf7348a0e3c3c1abfe162c5cbe2ac84", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96a072798e9d67f8d40be878850be1eb805fda1d6dd799194c15278760ab29ab", "generation_seconds_batch": 16.614415446063504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4040744304656982, "target_z": 0.7187500397630445, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.7611060738563538, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/88b57c6ceef1c14d00ab636f61f0649cdbf7348a0e3c3c1abfe162c5cbe2ac84.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "88cfc9bc2be31b74ba82b881d710b32c9fcec86f14d704899b8947aa7721d4b1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6752d74134aae6bc0ded3c0aade792c40b18c4498fe90f8a7ac9014064a7f92e", "generation_seconds_batch": 23.751342613017187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 6.221180438995361, "genuineness": 3.9425034523010254, "identity": 0.805935800075531, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/88cfc9bc2be31b74ba82b881d710b32c9fcec86f14d704899b8947aa7721d4b1.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "8957a4526c1a10db7dc5e1d57865b1da87693f374849d6e52f1a714d2d0042e1", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aaa0134541b40eee6cc46020820062d21c70bd5c808c8fd97de8e6f2641aa76d", "generation_seconds_batch": 14.748581159044988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 6.23706579208374, "genuineness": 4.118232727050781, "identity": 0.6041107177734375, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8957a4526c1a10db7dc5e1d57865b1da87693f374849d6e52f1a714d2d0042e1.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "8963b46ce6fa8cfa79e3d4d925361ba3e8f5887c3e54b21b1d472c0ad1c9266a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d40fd3dc9300c2b661fa3c58ded50c7750d8df8d583c926aae1be0ca0ae4af4f", "generation_seconds_batch": 25.169362952001393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.205295085906982, "genuineness": 4.21237325668335, "identity": 0.7599509954452515, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8963b46ce6fa8cfa79e3d4d925361ba3e8f5887c3e54b21b1d472c0ad1c9266a.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "897be4d7cdee2674509cd2573711a5ef13fc113b984a155d2b9ed0f34c5b689e", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa9cef5c583ab71e1f0d471d22857ed3103772e9bf093c12eb53f72aca852880", "generation_seconds_batch": 16.949406045954674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.467403411865234, "genuineness": 3.9801597595214844, "identity": 0.5947421193122864, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/897be4d7cdee2674509cd2573711a5ef13fc113b984a155d2b9ed0f34c5b689e.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "8a3ec6caa01cdc9ee68a83cd4336e54d74b3f6f3296e0ab2c70fe6d8c063d124", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8748ed428567e4b04265b826981fe947e32495e52a59abe47cbdfebd17df003", "generation_seconds_batch": 25.51077109691687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.7674742341041565, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8a3ec6caa01cdc9ee68a83cd4336e54d74b3f6f3296e0ab2c70fe6d8c063d124.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "8a860227d8a4dd6ec99d87931ae2c61f528e973430539912364a4cb344f6e2e1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26c6da10ca4dd3d4445cec40bfdb315d9ff51258ae349a75606054700c3b3e3e", "generation_seconds_batch": 21.637400410021655, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 6.515059471130371, "genuineness": 4.20609712600708, "identity": 0.47712957859039307, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8a860227d8a4dd6ec99d87931ae2c61f528e973430539912364a4cb344f6e2e1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "8a86d554fd6c5fbb7b610a6309b9f7665b3dc7a6fbfbecca973a2077e20f3432", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "718bab20da0240aa3828177b8d6dc5b64552794815a77c563be5b17cf1516e79", "generation_seconds_batch": 18.758449306013063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.205295085906982, "genuineness": 4.231201171875, "identity": 0.4847620725631714, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8a86d554fd6c5fbb7b610a6309b9f7665b3dc7a6fbfbecca973a2077e20f3432.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "8ad903e26a320b21710fafc6a75f7e771457498f47b7e860467f87db5514bdd0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80615d40ab1dcff7e73bd55a584ed1b65be4044dbc08aa8533e9e8b28344eb34", "generation_seconds_batch": 23.838195130927488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.535459280014038, "target_z": 2.468749966354347, "ce": 6.24500846862793, "genuineness": 4.093128204345703, "identity": 0.4542776644229889, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8ad903e26a320b21710fafc6a75f7e771457498f47b7e860467f87db5514bdd0.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "8af1abe6bc4ad4d96e05728f847f557d8edeb1f1ee5bc6761dcc402dbcfec243", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fbaba46058408692b0d1c27e2e8f5e34b8683ef185bf55be336b9af21c266b2", "generation_seconds_batch": 18.218167974962853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.32612681388855, "target_z": 2.296874952590216, "ce": 5.998785495758057, "genuineness": 3.7040140628814697, "identity": 0.6490350961685181, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8af1abe6bc4ad4d96e05728f847f557d8edeb1f1ee5bc6761dcc402dbcfec243.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "8b2840c2561053ec9f9eb6918a498f0acde468d1e6fa2e79ac2749c51483d966", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36c73cdae3455f5cac0c769461cd30e414295d70cb0a2f495be6cdd25fcee373", "generation_seconds_batch": 16.229206256917678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.4515180587768555, "genuineness": 4.3504462242126465, "identity": 0.3159492611885071, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8b2840c2561053ec9f9eb6918a498f0acde468d1e6fa2e79ac2749c51483d966.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "8b6c114759425b7dd21ada22cf4c3917255a67463553d4706abb3e8193e1b906", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cf9470ec99d74e7e65a5972f66555e3db96f006ad1501faa69ad0ef93b525651", "generation_seconds_batch": 17.769579211017117, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.802795648574829, "target_z": 1.867187418179889, "ce": 5.974957466125488, "genuineness": 4.337893962860107, "identity": 0.6706823110580444, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8b6c114759425b7dd21ada22cf4c3917255a67463553d4706abb3e8193e1b906.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "8bae4c5f3cd0f2b46ecc4ddbb9c119cc7a283fd84dc07aa024a67656b11806b5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21ed28c82936b3f6926ad78e85eeec9b57f3953c89d556a97e495ef842b85a89", "generation_seconds_batch": 23.21134830499068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5554027557373047, "target_z": 1.6640624197092369, "ce": 6.086154937744141, "genuineness": 4.193545341491699, "identity": 0.5400643944740295, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8bae4c5f3cd0f2b46ecc4ddbb9c119cc7a283fd84dc07aa024a67656b11806b5.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "8c2b8b9901b679adbb3db18ce1ad784dc5e7dcdcfd7b7cd3feb3ae8403ee4746", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d89dd43de626c00e19dd862022954db85c1e5ed3f175ba73e3d5bdf33d0c62c", "generation_seconds_batch": 22.64533573400695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.7113268971443176, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8c2b8b9901b679adbb3db18ce1ad784dc5e7dcdcfd7b7cd3feb3ae8403ee4746.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "8ca13a1b89ea8fbfbd7f4b7f5cdc87f3ce08823b2d4a30476aa2471b6149963a", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccd02d320a94b3c1d592f136422976e3bdbea572b7fe942bebc3319b8c21a413", "generation_seconds_batch": 16.627347600995563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.7371323704719543, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8ca13a1b89ea8fbfbd7f4b7f5cdc87f3ce08823b2d4a30476aa2471b6149963a.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "8cc09740936a76368125c84e7f8dc698ac3b46412946cb81023e90bfcf1d68aa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60ea3e6b3ba45a44df2a1e4a2ee00b03b75c535923671272de14948857caa284", "generation_seconds_batch": 16.904202364035882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 5.8875885009765625, "genuineness": 3.8044304847717285, "identity": 0.6844708919525146, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8cc09740936a76368125c84e7f8dc698ac3b46412946cb81023e90bfcf1d68aa.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "8cceb0c8cd58a8b0cd57fdfc20ca3dce0d663e2893aefd87506dd5ead89dd698", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bffab9c34acf64b8bd90b32538ad6c739778da47a8be3a631021e8cbe0a1dea", "generation_seconds_batch": 14.655249343952164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.0146708488464355, "genuineness": 4.18726921081543, "identity": 0.6994974613189697, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8cceb0c8cd58a8b0cd57fdfc20ca3dce0d663e2893aefd87506dd5ead89dd698.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "8d3124894a5a40e75472044c56c59440e1358f8dfdba667ca3abdef470f7983e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e30c34fe60caf0a2e98101ff5b1b0264ba5263dc49afff7bca0cb9166788ed3", "generation_seconds_batch": 23.95283246494364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.260893821716309, "genuineness": 4.218649387359619, "identity": 0.6497384905815125, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8d3124894a5a40e75472044c56c59440e1358f8dfdba667ca3abdef470f7983e.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "8d3c4cc7d8d570b955b0f93775ba0a59faf9364cc5c6b0a55d0d29ec8ac1ea65", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e565dcb3f78d57413e77bfc767ac3ad0032845c96da4a95969b0be7a848eaf6b", "generation_seconds_batch": 16.52818385302089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9178904294967651, "target_z": 1.1406250290576094, "ce": 6.252951145172119, "genuineness": 3.8295347690582275, "identity": 0.349571168422699, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8d3c4cc7d8d570b955b0f93775ba0a59faf9364cc5c6b0a55d0d29ec8ac1ea65.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "8de359ae59b79463576b8e55111baaccf1881f0130a48afd9a110ce027af7251", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb878c86595dbd05f170c9fd23e9b2d84e14f41e951146d0ce85bb52e056538c", "generation_seconds_batch": 25.163893157965504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.832254409790039, "target_z": 1.0703125145288048, "ce": 5.935244083404541, "genuineness": 4.20609712600708, "identity": 0.613895058631897, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8de359ae59b79463576b8e55111baaccf1881f0130a48afd9a110ce027af7251.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "8e08f989025f05909550f76327d5bdf716685a497fb3c02cfee6b9db66d8790f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a74a31a2be8698631fa0440bdde52c74bf95e7b477a04d451d10150fefd1c57c", "generation_seconds_batch": 26.64413145906292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 6.4912309646606445, "genuineness": 4.457139015197754, "identity": 0.35698527097702026, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8e08f989025f05909550f76327d5bdf716685a497fb3c02cfee6b9db66d8790f.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "8e6424a10c9d1ec2fc77b81502a34f365dedb6a91e6627c716a2ceba430b7a83", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aeaa77f728f674faf0b63ca7c84d985f0a5de725648c0b5285c43ef169dce393", "generation_seconds_batch": 16.750297656049952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 5.967014789581299, "genuineness": 4.174716949462891, "identity": 0.6993863582611084, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8e6424a10c9d1ec2fc77b81502a34f365dedb6a91e6627c716a2ceba430b7a83.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "8e66953a9686c52780b4d638c7553770db503c464f6980307061f8a3a544115d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffeb88d9fbc8c08534ecebd6101bff2df4286bab8228e88994e638571bb4a02d", "generation_seconds_batch": 15.944371436955407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7837655544281006, "target_z": 1.851562523704892, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.7207443714141846, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8e66953a9686c52780b4d638c7553770db503c464f6980307061f8a3a544115d.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "8e7b70f29e5cac13d58d939889cfbef40aab9680c73cccb9c9ae29142605fe48", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20c1c96153f521ccfee586dd39cf0a23aff6904c004c59afa971ef47e76c1683", "generation_seconds_batch": 22.675465599983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4897103309631348, "target_z": 0.7890624564135857, "ce": 6.38003396987915, "genuineness": 4.293961524963379, "identity": 0.6431776881217957, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8e7b70f29e5cac13d58d939889cfbef40aab9680c73cccb9c9ae29142605fe48.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "8ebe5c9b5648e9c0c210defa92f5a14f9e3ea216b6c60f3e40249f393c424666", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15ba1054f2a67c415aa34751c5038da70f4c865ff5a03f998216a4f176ca0b2d", "generation_seconds_batch": 17.397432690951973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.173524379730225, "genuineness": 3.8922953605651855, "identity": 0.21361958980560303, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8ebe5c9b5648e9c0c210defa92f5a14f9e3ea216b6c60f3e40249f393c424666.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "8effc5a8c2a2774d9ef84740a5a1d2097a1adea82c268eb070177ca1b6786d7b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66e2e99891a4cdeea8d45a51be533f611d15725fc878a2fcca7e880193b8ccce", "generation_seconds_batch": 22.87776218703948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 5.911416053771973, "genuineness": 3.848362922668457, "identity": 0.5883446335792542, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8effc5a8c2a2774d9ef84740a5a1d2097a1adea82c268eb070177ca1b6786d7b.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "8f8ece62e3670f79be312c408ada6f9a8efc14d5dce28db7e415729b91dfdefd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "696f5abaf3b5e5d585cd96abee5dd01d160924af6090740aab475549b8aca74c", "generation_seconds_batch": 25.816367791965604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.726674795150757, "target_z": 1.8046874487668465, "ce": 6.046441555023193, "genuineness": 4.268857479095459, "identity": 0.7243959903717041, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8f8ece62e3670f79be312c408ada6f9a8efc14d5dce28db7e415729b91dfdefd.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "8f9f64c120bf60b88a44a8457280158fed28faaad20de2121cd6fb019a82acec", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ecf261a8829a9af26e3f21dbc6c8265b31ff023f45feb0070081711d3c7c45c", "generation_seconds_batch": 16.52818385302089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.340320587158203, "genuineness": 3.7040140628814697, "identity": 0.32307732105255127, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8f9f64c120bf60b88a44a8457280158fed28faaad20de2121cd6fb019a82acec.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "8fa623a8115340ceb079c48b6934d075a7c05890303c1fbe9a71419ffb48f933", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9956121b1f48bc490d167620c6aeffc2aab85bd9872cee5b285edb4f8ef7e950", "generation_seconds_batch": 19.062100211973302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.062326908111572, "genuineness": 4.15588903427124, "identity": 0.48729458451271057, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8fa623a8115340ceb079c48b6934d075a7c05890303c1fbe9a71419ffb48f933.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "8fae12a5e0e8c24a5e5674d9f46fa8ab20e4e2bc5aa7cb5df808956b71b277ac", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fbbaf75759f3ca33f5fb9452b64e91fb1df18769bc126e9ff924e76f5eb87248", "generation_seconds_batch": 23.754654710995965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.260893821716309, "genuineness": 4.143336772918701, "identity": 0.6342058777809143, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/8fae12a5e0e8c24a5e5674d9f46fa8ab20e4e2bc5aa7cb5df808956b71b277ac.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "906feb7e47d7c4a95a32252ab1a9628d1ad849f1152544cb89c46c0614845d4d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "044f1cc7a1ddcec278a645821e28d91b2cff0ec7624181ad388bfdee1aef719b", "generation_seconds_batch": 16.993433407042176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.157639026641846, "genuineness": 3.9801597595214844, "identity": 0.664912223815918, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/906feb7e47d7c4a95a32252ab1a9628d1ad849f1152544cb89c46c0614845d4d.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "90a6971e3647e67eab2e9db495967d6c11243ce883a46952e53fcb27390a43d6", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48255f6695feb1ad29be625c9508bdfcca7267dbd7a1fce5a1bfd331829f617d", "generation_seconds_batch": 17.864783053984866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 5.895530700683594, "genuineness": 3.8546390533447266, "identity": 0.6454294919967651, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/90a6971e3647e67eab2e9db495967d6c11243ce883a46952e53fcb27390a43d6.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "90afa163470b93ce6bc3145ba203ab2994af77830da54bd969e3f51eb78aed2e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a71c20503522bdf0e74562cad54e818330d86c9a2eee1140ea56bbc8c3385c8", "generation_seconds_batch": 21.83630969701335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3460702896118164, "target_z": 1.4921874059451061, "ce": 6.09409761428833, "genuineness": 4.425758361816406, "identity": 0.6911603212356567, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/90afa163470b93ce6bc3145ba203ab2994af77830da54bd969e3f51eb78aed2e.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "91ade57052d2d77e365d79ff90149445c4a337e52ae1936c97df6a1a0e6b70fe", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fd701f5ce10329bf3d832ab49bcca2d97e422e1f92df39b8373d8c5769cff3d", "generation_seconds_batch": 17.63052460399922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.038498878479004, "genuineness": 4.419482707977295, "identity": 0.5273568630218506, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/91ade57052d2d77e365d79ff90149445c4a337e52ae1936c97df6a1a0e6b70fe.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "91e3fd8ca993ae94b4b1232267197bb1d77af510f742cc8e8964c22db8c3013e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bfd3bc1b4968e522ad6116db366378ddfe2436114a62bb6e961c66a86072103", "generation_seconds_batch": 27.53396629903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.546830177307129, "genuineness": 4.15588903427124, "identity": 0.5805690288543701, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/91e3fd8ca993ae94b4b1232267197bb1d77af510f742cc8e8964c22db8c3013e.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "9203ad4557940bd20dca3b4f6418008dcfac331bddbf68ebf58a82ae0344db84", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eb3e438e5d1c4a29edc7c2e3f8b729ef4931f5b12aed3d296baf5cca876587c", "generation_seconds_batch": 23.47612225904595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.070269584655762, "genuineness": 3.835810661315918, "identity": 0.7149532437324524, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9203ad4557940bd20dca3b4f6418008dcfac331bddbf68ebf58a82ae0344db84.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "920ac2f834d661e000f2fddba599f4ebf36644a8b1f69995b424131edd8e818c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e127d1d81d8aff2c286a0a8eb015f5a742f6aec0ae3df9bf8acc5d9642e9eecd", "generation_seconds_batch": 22.92463071702514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.09409761428833, "genuineness": 3.9550557136535645, "identity": 0.40593963861465454, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/920ac2f834d661e000f2fddba599f4ebf36644a8b1f69995b424131edd8e818c.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "923980a49e3f428324fb4f08fefb3acee026cf7999fac1452d08ef8525a7231f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dce2a306394f2bf10a25347bd855de1916b8c2c5198ccf46c225a00a58667ce1", "generation_seconds_batch": 23.751342613017187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 5.903473377227783, "genuineness": 3.9299514293670654, "identity": 0.6641699075698853, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/923980a49e3f428324fb4f08fefb3acee026cf7999fac1452d08ef8525a7231f.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "92917273f90313e3a75a2257f3992d6a0cc23a94655217e5fd925a5df2bfd673", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1650dfd8da9299c90de22fb5b2915e24d6943dcdf109c74621b6cd6898cae471", "generation_seconds_batch": 23.142422560951672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 5.998785495758057, "genuineness": 4.174716949462891, "identity": 0.7874511480331421, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/92917273f90313e3a75a2257f3992d6a0cc23a94655217e5fd925a5df2bfd673.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "92922c2656af6e2fcb0412e8bba3aecfb31e7ef59999279cffe2fb0523fefb33", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64bfdf1f804af90b8df69f2617760ada31ee65802711e2ef99c8c0e90d9dca5e", "generation_seconds_batch": 20.430139241972938, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.292664527893066, "genuineness": 4.118232727050781, "identity": 0.7904287576675415, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/92922c2656af6e2fcb0412e8bba3aecfb31e7ef59999279cffe2fb0523fefb33.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "92b6eae12b10abf82bdc112b996b99cd441980da8cabb94ea84efde5f9e2fede", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdbef57d2c12cdf11035dc272a6529dee8c43da22f1d18a5d2c4f189056eef62", "generation_seconds_batch": 17.415100169950165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.5217826962471008, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/92b6eae12b10abf82bdc112b996b99cd441980da8cabb94ea84efde5f9e2fede.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "930ace9f43efc23a866127411f9b35e972fd1efd7e9ed4d240b612d96cd47f6c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "992d9851f4559869b0164d757a7e341ab3ef2686cd5c4cbdd96aef323ac0d460", "generation_seconds_batch": 24.355041628936306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.475345611572266, "genuineness": 4.193545341491699, "identity": 0.6652708649635315, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/930ace9f43efc23a866127411f9b35e972fd1efd7e9ed4d240b612d96cd47f6c.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "930dcedfee4664428fa0ee8f8752b49de6a615fc03b319e6646e91c80ae804cb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "468ab1aa9abb2d575c2d2cdd19b529dd0a903b3de1ef432c77d02b5a6303dd89", "generation_seconds_batch": 15.39561661996413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.141753673553467, "genuineness": 3.7542223930358887, "identity": 0.3485516309738159, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/930dcedfee4664428fa0ee8f8752b49de6a615fc03b319e6646e91c80ae804cb.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "9321a2b4d2c3dee46e4f326ca259d998cc448ae5d2198d6661fab80a72d3be5d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ea650a2420d52a796d096f98c4cb3a3c54dd8d096baf5eb070bd719bd02b1cb", "generation_seconds_batch": 19.039650380029343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.173524379730225, "genuineness": 4.507347106933594, "identity": 0.6635007858276367, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9321a2b4d2c3dee46e4f326ca259d998cc448ae5d2198d6661fab80a72d3be5d.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "93525b0deaef6b5b77cac822bf91bafb9a9d716db97a6e377405b1519ed6f970", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc8710e89cb47f9bb4d4ca0fa1a0853d721df0419a19154f27e16d920cdbffc9", "generation_seconds_batch": 14.185862428043038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 5.903473377227783, "genuineness": 4.281409740447998, "identity": 0.6550877094268799, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/93525b0deaef6b5b77cac822bf91bafb9a9d716db97a6e377405b1519ed6f970.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "937a28364e94c3ee5ef504c4e6bf01d56bff7b5794f5e0e54d2077011ee90d07", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a4ed03cca6ffd04ac34e8144f09409d1752a68b7813601f037897836820019e", "generation_seconds_batch": 21.734900022041984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.070269584655762, "genuineness": 3.9173994064331055, "identity": 0.5330384969711304, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/937a28364e94c3ee5ef504c4e6bf01d56bff7b5794f5e0e54d2077011ee90d07.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "93bd50eee95b7744a4faff2f23ab4a1986b3db43621e4ecf8eda65d715363ed6", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3776188cd22684295e8f685de9ceb9effbdd8f9062005a53bf45e23c56b6cd44", "generation_seconds_batch": 19.46838003094308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.141753673553467, "genuineness": 3.7416701316833496, "identity": 0.5184711217880249, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/93bd50eee95b7744a4faff2f23ab4a1986b3db43621e4ecf8eda65d715363ed6.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "93fef4a08682886d2bd128a529bbacc14a95534647788d26147d2324924ddd0f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1df517ed4642f4de2ee703064cd70f76c22a505e5bc4cf8a5489f3e808ae255c", "generation_seconds_batch": 16.103974448982626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.7944707870483398, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/93fef4a08682886d2bd128a529bbacc14a95534647788d26147d2324924ddd0f.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "94006892a9b04c829190528c9c43ecd1a9cee8b76fcf37b765502d0383b599b7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b14a4d89e0daf2842c384830a8ba1a7f9f30321e31231fdc58d660e3670c61cc", "generation_seconds_batch": 23.88074375106953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.078212261199951, "genuineness": 3.8797430992126465, "identity": 0.18299870193004608, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94006892a9b04c829190528c9c43ecd1a9cee8b76fcf37b765502d0383b599b7.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "9402ab6d18aeaa8e9dc60cd69c1fef28c46fa17cec8d820b7fe7e64226281bdc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac05d3c60e904c818138d3509a48d8dd7887737476d701807f52992cb081619c", "generation_seconds_batch": 22.17268229695037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.332377910614014, "genuineness": 4.193545341491699, "identity": 0.1824447214603424, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9402ab6d18aeaa8e9dc60cd69c1fef28c46fa17cec8d820b7fe7e64226281bdc.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "94120bb9f000f28cfb63dcb1369d2087b4364a60bd29518e936a94f37c119350", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6dc570f5fa10eee9a7f29a4e1f8ef574502d692279e47dd43781b3668d174f00", "generation_seconds_batch": 16.00133712904062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.348263263702393, "genuineness": 4.482243061065674, "identity": 0.42787009477615356, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94120bb9f000f28cfb63dcb1369d2087b4364a60bd29518e936a94f37c119350.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "94634d8aad335b88596fe5f5e95cf1878062506041a8b92decd230363b79b3a0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba59d69f4e43afc7920814b289c928b56427cebc1de2f2609ef7bccafdc1aa9a", "generation_seconds_batch": 17.206857195007615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 5.967014789581299, "genuineness": 4.18099308013916, "identity": 0.567035436630249, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94634d8aad335b88596fe5f5e95cf1878062506041a8b92decd230363b79b3a0.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "94706082bdb54a9c1e3f452aeafe13dd3c7abbd24c77271754b9349686951379", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9e58e1dedbedf8ff68532549dcfbcbf7366ae7dc46efff3e43f73bea944b19be", "generation_seconds_batch": 25.922847214038484, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.897946834564209, "target_z": 1.945312477824456, "ce": 6.165581703186035, "genuineness": 4.3127899169921875, "identity": 0.7316899299621582, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94706082bdb54a9c1e3f452aeafe13dd3c7abbd24c77271754b9349686951379.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "94b7dca1393c46ee1ea5ad7d736df165a11537509697d48e0685e73501c93081", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e2227c4f2c4d316fd9c717c9bc564f648eec3de359149c4d629f3f4aa8ba526", "generation_seconds_batch": 26.883209926076233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.7503259778022766, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Fanfouilly, faire Dan Messidas et", "wer": 0.2777777777777778, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94b7dca1393c46ee1ea5ad7d736df165a11537509697d48e0685e73501c93081.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "94e5f2cd82b38402f714ee243df500fc10c8e671d0b04e788ae9aa1804b58311", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e54418670ac8aaed294e7ba48d384ad5b9765d68e73e9064320041ac110c91a0", "generation_seconds_batch": 25.266535451984964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.062326908111572, "genuineness": 4.030367851257324, "identity": 0.7416489720344543, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94e5f2cd82b38402f714ee243df500fc10c8e671d0b04e788ae9aa1804b58311.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "94f84a4d7eac17d0adc1d616800c088c8de62e98f987f9b8475d66c8ce44b182", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "861305dd445f3476210cf14ce361428a2163ef32fb731cdf06c54f7609319ee3", "generation_seconds_batch": 25.002348406007513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886143684387207, "target_z": 1.7734374640603252, "ce": 6.173524379730225, "genuineness": 3.8922953605651855, "identity": 0.3319648504257202, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/94f84a4d7eac17d0adc1d616800c088c8de62e98f987f9b8475d66c8ce44b182.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "9542514b3192bdc761bb4edf6462f35ab398fe1bb1ec80bc29997722223e8735", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82b9a866cd5985c154e30befd336b6c6b528fb9a621beff9c1d11e836adf2637", "generation_seconds_batch": 22.51298378407955, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.49831223487854, "target_z": 1.6171875405277185, "ce": 6.23706579208374, "genuineness": 3.625563621520996, "identity": 0.42317700386047363, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9542514b3192bdc761bb4edf6462f35ab398fe1bb1ec80bc29997722223e8735.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "95c6afe449f01350efb9b87c49634facb705ed4f306f2ee16382bcd274f7dba5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4fc92ae65d7dc1590749bb6f7519adaddb3559cc46ee22ef36d0a968c3c5c66e", "generation_seconds_batch": 23.735393090057187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.372091293334961, "genuineness": 4.243753433227539, "identity": 0.502745509147644, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/95c6afe449f01350efb9b87c49634facb705ed4f306f2ee16382bcd274f7dba5.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "95d1a90d638accd2084b39045a4e37d6bf43e0dbbc6dbb228aef706fceae37c5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffcee6a8516d8d2a5c6b39060c06ff296d9ea91ebf065f3032ef0e977835859f", "generation_seconds_batch": 23.034495058003813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 5.998785495758057, "genuineness": 4.268857479095459, "identity": 0.6492748260498047, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/95d1a90d638accd2084b39045a4e37d6bf43e0dbbc6dbb228aef706fceae37c5.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "9642b1af41b4d748f99a1e0ae9523366fa2b7038504e723d8ee82b9c19fb40bb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1be10b1c7d4e74b5740e521eddb1e69c164790dc8e8806781a4179dea907439", "generation_seconds_batch": 24.601513013942167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.764735221862793, "target_z": 1.8359374334733678, "ce": 6.157639026641846, "genuineness": 3.7667744159698486, "identity": 0.2615220546722412, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9642b1af41b4d748f99a1e0ae9523366fa2b7038504e723d8ee82b9c19fb40bb.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "968f98214099154a7214d5fe3be9033075efc6e342353a3e0a5397740dc5df36", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05154ce39477143a35a0a40e77e7de5bb383f3d73dfea3725fd96b34693fe8c9", "generation_seconds_batch": 23.24653733393643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.3645181655883789, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/968f98214099154a7214d5fe3be9033075efc6e342353a3e0a5397740dc5df36.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "96c0f117bfd083f054e765eb625b7b4995a10fc8d54df04e961437e3d10ed779", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea25529850cb9443aeb02666710327dae573ea24cae85d6ff51df3b6c4a4b5e8", "generation_seconds_batch": 22.1711235078983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.284721851348877, "genuineness": 4.105680465698242, "identity": 0.05778106302022934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/96c0f117bfd083f054e765eb625b7b4995a10fc8d54df04e961437e3d10ed779.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "972bff967c125e531b06a9899a3c41b235a8f149d73de29697adb2b58bf988b5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e7145ff2a69a61a728c01628d58932a4bd0f46bf21b9e6a7bd7ee8f2683e4c7", "generation_seconds_batch": 17.078591127996333, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.070269584655762, "genuineness": 3.9173994064331055, "identity": 0.7255215644836426, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/972bff967c125e531b06a9899a3c41b235a8f149d73de29697adb2b58bf988b5.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "973586421ea9c9d604bd410ad0d1a81bbb55227b39c9c11824e8af27b990e8e0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e457a4db5d9ba726e704c6e42620ac9a850b2b7a9f6fd386932bc63743c0bf37", "generation_seconds_batch": 23.11167255300097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.181467056274414, "genuineness": 4.124508857727051, "identity": 0.6267681121826172, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/973586421ea9c9d604bd410ad0d1a81bbb55227b39c9c11824e8af27b990e8e0.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "9739a1f6caec4d4d6142ad419dca0bc60544e713218273777fa79e845610f5bd", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41b95d83bd4362afe7fa29bf850031cdffa646fece5f0ec5d761d1dc019a0d0a", "generation_seconds_batch": 17.287164672045037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7751637697219849, "target_z": 1.0234375374690228, "ce": 6.308549880981445, "genuineness": 4.149612903594971, "identity": 0.5141552686691284, "hypothesis": "Deine Nachricht kam, als sie schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9739a1f6caec4d4d6142ad419dca0bc60544e713218273777fa79e845610f5bd.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "980315e8dd0719ff607224c15442a3935f0391b15a3396c8ae9e62049e31fbf9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5ff7064e80bf937c63b0a7bec0186aeac67333b93bcfa59c459ace9a6546c1c", "generation_seconds_batch": 16.805235796025954, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 6.157639026641846, "genuineness": 4.557555198669434, "identity": 0.7260650992393494, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/980315e8dd0719ff607224c15442a3935f0391b15a3396c8ae9e62049e31fbf9.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "9872d035c4f675a0d96419b108ee54ab082451af48cee9b993d0c8575c3768f7", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e79067c09f1069a737d70ebefa231619a5bc0b1bc9f0d5695832852124119e06", "generation_seconds_batch": 16.31849658500869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.40316104888916, "target_z": 1.5390624808831517, "ce": 6.24500846862793, "genuineness": 3.8797430992126465, "identity": 0.28485557436943054, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9872d035c4f675a0d96419b108ee54ab082451af48cee9b993d0c8575c3768f7.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "989b6a0ae6b3743176c0eabc167b94b641b6b4387903f28625557d3c7b162693", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d5abcc833c54046f86559116d555c9ee573287291e589e787e803a5e48a94ea", "generation_seconds_batch": 16.547664126963355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.5406864881515503, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/989b6a0ae6b3743176c0eabc167b94b641b6b4387903f28625557d3c7b162693.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "995bbe0a14fcd3d4eeb28d5b4fc7de72b3e72ea85afd1fc2ab9d75acaf9c36fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa8a5f403f91c86b9132a2c754d01347d8086b2888c2f1426a0e466f41ba83f0", "generation_seconds_batch": 26.311546046985313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.340320587158203, "genuineness": 4.369274139404297, "identity": 0.509445071220398, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/995bbe0a14fcd3d4eeb28d5b4fc7de72b3e72ea85afd1fc2ab9d75acaf9c36fa.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "99746309358872bbb73d787213795cd05db007445479a98a4a71c79188ba7c94", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f282f00ff0f9429f9c80c0418279361a2ae3e87a8db3deee8f5ae47abbcc7dc", "generation_seconds_batch": 21.040646870969795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.109982967376709, "genuineness": 4.2751336097717285, "identity": 0.7799602746963501, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/99746309358872bbb73d787213795cd05db007445479a98a4a71c79188ba7c94.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08aa02d70f20ab47426767ac9fa3dd61c103b8c5bcdd8a4c32098a5f8fa7ef3d", "generation_seconds_batch": 17.0768135729013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3945592641830444, "target_z": 0.7109374946472825, "ce": 6.403861999511719, "genuineness": 4.532451152801514, "identity": 0.736314594745636, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "9a030b9f5e5f78daaa53feb55c8aba3cea34d2e7d51b72949cf41a866eacf3f5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "646197e1a3ee59092ddd94083dfe64b72c37bdef3fcf9f3efb39ef7b89e009e0", "generation_seconds_batch": 17.6602623880608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.098677635192871, "target_z": 1.289062603230981, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.5762280225753784, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9a030b9f5e5f78daaa53feb55c8aba3cea34d2e7d51b72949cf41a866eacf3f5.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9a1db9263484824cdec4b166e586f0783aeeae17432b77e4fd5421c84ad413f3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0051cf6ef96030228e4bd7a1acd9109c27171432b659e1f0fbf1fef7d54e844", "generation_seconds_batch": 23.25263145903591, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.427689552307129, "genuineness": 4.362998008728027, "identity": 0.19967402517795563, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9a1db9263484824cdec4b166e586f0783aeeae17432b77e4fd5421c84ad413f3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "9a226d372ca0995cec9aaf00eebb4747df88cca2e3fdab7642bbab260fe4e006", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da90c8396a0375724f96423b89070a3d51588f2dd71efd10fc0f181605e311ab", "generation_seconds_batch": 24.047121389070526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6562248468399048, "target_z": 0.925781261852446, "ce": 6.117925643920898, "genuineness": 4.607763767242432, "identity": 0.7912350296974182, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9a226d372ca0995cec9aaf00eebb4747df88cca2e3fdab7642bbab260fe4e006.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9a4a73ed4b29964fe619aea5e77bb3f47f14f731f7cc56b24b7176bcbb3daa7e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8aa6a1997e0b95cb4ca4752cf49e3bb2ba2acef5d3bd6ff9c0c5b2d07bf21de4", "generation_seconds_batch": 23.54422197304666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.308549880981445, "genuineness": 3.7416701316833496, "identity": 0.1850913166999817, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9a4a73ed4b29964fe619aea5e77bb3f47f14f731f7cc56b24b7176bcbb3daa7e.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9ab4651c648e4ab9725d478f9b803d52500e8092071b7f22b02cc228e0d41ac2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8be2e127a96c41c260e457da1b2d0c5a00021e0926ae606873c951a2ac656094", "generation_seconds_batch": 17.32417581800837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.5382564663887024, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9ab4651c648e4ab9725d478f9b803d52500e8092071b7f22b02cc228e0d41ac2.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "9aba877fb4958289c6eae3b32b7e29cb2f2a9e12322b910aa18d23a6821401f6", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "637ddcadd6ffba39b762b91f9a96aec6f6b3bbc22f1485b44e336be720554943", "generation_seconds_batch": 15.949925204040483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.340320587158203, "genuineness": 4.036643981933594, "identity": 0.2602013349533081, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9aba877fb4958289c6eae3b32b7e29cb2f2a9e12322b910aa18d23a6821401f6.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9b2f1a37d22e130d278496e36faeaa696c84493ab162cfa13c42d67c2939d095", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "410a58136cbe06795506c160cd71c2ee0577e4a69028158c00b2e6ffd4c99bfb", "generation_seconds_batch": 18.089015537058003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.3552740812301636, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreuen.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9b2f1a37d22e130d278496e36faeaa696c84493ab162cfa13c42d67c2939d095.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "9b48adf495f83071b569a09c1109c9b3f02ee8803808e0d77004086bf59a3961", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9fce3edb7c45e14686f47d1ddde292c9697893ca64cfdbcea165c14b44fab5f", "generation_seconds_batch": 24.948357441928238, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886143684387207, "target_z": 1.7734374640603252, "ce": 6.24500846862793, "genuineness": 4.281409740447998, "identity": 0.6467570662498474, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9b48adf495f83071b569a09c1109c9b3f02ee8803808e0d77004086bf59a3961.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "9b877c5e8236cdef6e71d6d4354ddafd7f6269a67463bb4299ada72aa05915aa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91043a7e1e2843a6c77f0f780f1b0a54870a3bfde20d5e61eb890be89417e1fe", "generation_seconds_batch": 24.809131453977898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.046441555023193, "genuineness": 4.042920112609863, "identity": 0.45938166975975037, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9b877c5e8236cdef6e71d6d4354ddafd7f6269a67463bb4299ada72aa05915aa.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "9bfb6e1630fa9f0efb4e8d6d75895604a3235616d9ab32030797a26c27f96391", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fd91315b696b6c36f1a8f1f7620f0ab0f1ddce71f40b91399c925553b798a0d", "generation_seconds_batch": 21.941196984960698, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.6880984306335449, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9bfb6e1630fa9f0efb4e8d6d75895604a3235616d9ab32030797a26c27f96391.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "766ee7d47d20a3825eac85d06975cd2f7e7c89d9849ddc73b6b0cfef245059d8", "generation_seconds_batch": 17.068734186934307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.181467056274414, "genuineness": 4.406930446624756, "identity": 0.7015600204467773, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "9c4b8ace429860679c04e35bcebf4bfc3606ae9845147c525377187704b5975b", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c4852c229a19c51b6a3fa31f3390c76dd08a3475314d123d7d9fe050126afab", "generation_seconds_batch": 26.58764007501304, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.284721851348877, "genuineness": 4.13078498840332, "identity": 0.6121469736099243, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9c4b8ace429860679c04e35bcebf4bfc3606ae9845147c525377187704b5975b.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "9c636d941f67004a1cddbbc94660b0dabfdaed1d3bf1cfd44f18065a54f7bda7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a59f0fb5bfe471a0d5844d1fee6e1708baab14478cc68538211b007cfcb9a11", "generation_seconds_batch": 24.624230314977467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.292664527893066, "genuineness": 4.2374773025512695, "identity": 0.2930179834365845, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9c636d941f67004a1cddbbc94660b0dabfdaed1d3bf1cfd44f18065a54f7bda7.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14c3407204b8a0cc1c84cc107820150e87bbc206e133c0e73f3ed49a7d6a39a2", "generation_seconds_batch": 17.21819578798022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886143684387207, "target_z": 1.7734374640603252, "ce": 6.038498878479004, "genuineness": 4.218649387359619, "identity": 0.6794692277908325, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "9cb44da53c50d8704f6eef53269297d7de15835e030f14ba40fbdc6005fb52b7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4389d3dc4ef456dcc9b7a267beffa8ad69df27544e31f75d791db3b65cec2dac", "generation_seconds_batch": 22.796518290997483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.292664527893066, "genuineness": 4.262581825256348, "identity": 0.6927878856658936, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9cb44da53c50d8704f6eef53269297d7de15835e030f14ba40fbdc6005fb52b7.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "9cbf12825f7307ff244f39fc9e0b10ad9a4ce8a6b851a9fcd417771811d6dd3d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05b6fc3042642c1f2a93251de3315488b38d7c9c283193411b5f1ebf5f686220", "generation_seconds_batch": 15.949925204040483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.38003396987915, "genuineness": 3.9048471450805664, "identity": 0.5201159715652466, "hypothesis": "The road turns left, behind the bridge. Keep going, until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9cbf12825f7307ff244f39fc9e0b10ad9a4ce8a6b851a9fcd417771811d6dd3d.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9d3df84902d25f2fc8ae6e6351c22115cb394031a55ee1d37066fd73a25c2f43", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56305bfe6ff413825ef4034e303761b1e7c59a45b392a0d4640e645fe67086cc", "generation_seconds_batch": 24.121734102023765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.403861999511719, "genuineness": 4.093128204345703, "identity": 0.5010669231414795, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9d3df84902d25f2fc8ae6e6351c22115cb394031a55ee1d37066fd73a25c2f43.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "9d48fcb2635083d627b05314cbf38acb4aa01ff9a2982101406c7d7cf3968531", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cb5a3acde88cb2ee1a2906ffcbd44ab4676554b8cdc5eebb00cac026ea88655", "generation_seconds_batch": 15.2062165789539, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.0305562019348145, "genuineness": 4.2374773025512695, "identity": 0.5939558148384094, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9d48fcb2635083d627b05314cbf38acb4aa01ff9a2982101406c7d7cf3968531.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "9d572fc1c7c4a25c4c5ad377383502c28703e506b151e06b55d7fd4deb1087d3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "098babea3b4c0b30d5843a355e0bf67f81fedd243ecfcab20198ccac9539d6cd", "generation_seconds_batch": 16.463702799985185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.213237762451172, "genuineness": 4.293961524963379, "identity": 0.7102780342102051, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9d572fc1c7c4a25c4c5ad377383502c28703e506b151e06b55d7fd4deb1087d3.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "9d59615139327b722efe82325ca49d318187ca4e47eb72fb757d749da92850bc", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d3a816bbbf57140d9b8485647f06b621defd717e906ad7ec78ecbe21cd374d1", "generation_seconds_batch": 22.78242627205327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.356205940246582, "genuineness": 4.381826400756836, "identity": 0.590501606464386, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9d59615139327b722efe82325ca49d318187ca4e47eb72fb757d749da92850bc.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "9d9f38c3c6261bd249d297efe5eeb3905e7ba0dfd421cbb062bc6e0b8ddaca88", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1d47d64560eeab4d69325905430531bcdf5ce583321cc378cb5fd32f8368248", "generation_seconds_batch": 16.386504685971886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.229123115539551, "genuineness": 4.256305694580078, "identity": 0.7321028709411621, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9d9f38c3c6261bd249d297efe5eeb3905e7ba0dfd421cbb062bc6e0b8ddaca88.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "9de5cd361f5a8981d064bb5297a60525f69e9f484ce52453a72f8419171d7384", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed00e31a0131a9d261293e9f864cf0172baa74b8529d5cc761f3ec2e2559c224", "generation_seconds_batch": 16.121033271076158, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.4189915060997009, "hypothesis": "Geht's, am Minkticht, die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.15789473684210525, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9de5cd361f5a8981d064bb5297a60525f69e9f484ce52453a72f8419171d7384.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "9e28d80a8ca9e898a7ab0ee0979af159d57a529748b4b3724cd2bd7ce2a369f1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abd9397bf37d2ed5ec4dd3adc51a7e642d13d7d95069d024482cd47b1a0f8e26", "generation_seconds_batch": 19.403771906974725, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.340320587158203, "genuineness": 4.325342178344727, "identity": 0.3541650176048279, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9e28d80a8ca9e898a7ab0ee0979af159d57a529748b4b3724cd2bd7ce2a369f1.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "9e664ed32009d3fcf2bad1df5b346dd141503c2423e3ba8b06d93c030eb23f1d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa56a1219a9968c38dce754cc8c2a07842dfff647aa1689f08369a70c4de01f1", "generation_seconds_batch": 22.80701476393733, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.260893821716309, "genuineness": 4.218649387359619, "identity": 0.5659729838371277, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9e664ed32009d3fcf2bad1df5b346dd141503c2423e3ba8b06d93c030eb23f1d.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "9eb2641c86a257321a969b86c54b45a65289c5f3fec7d2d07dff52aaa19c960b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fde316eb21f0a5a0b5e7c49456a17507e5991281d43612a38e0627904cd6f9ad", "generation_seconds_batch": 17.218335265060887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.221180438995361, "genuineness": 4.124508857727051, "identity": 0.41373223066329956, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9eb2641c86a257321a969b86c54b45a65289c5f3fec7d2d07dff52aaa19c960b.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9ebfcaa486a6e3783187220fc9e3fbc66b9085d811c6252c2d9e365e1eef2733", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4b9e22de1a4f1dd1718381036bf5b97c762f3304878d0adf9dc61799ff651e3", "generation_seconds_batch": 22.10417850909289, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.1894097328186035, "genuineness": 4.105680465698242, "identity": 0.28715944290161133, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9ebfcaa486a6e3783187220fc9e3fbc66b9085d811c6252c2d9e365e1eef2733.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "9ecf472408cf196a858e6300e4fac6250d18c425f714c4180e4d59072eb2f7f1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bcae04bf6f984bde33ee44fd8573ab90953ea4d93eb7c8252546a8e27adb235", "generation_seconds_batch": 26.094036417081952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 5.927301406860352, "genuineness": 4.406930446624756, "identity": 0.6068991422653198, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9ecf472408cf196a858e6300e4fac6250d18c425f714c4180e4d59072eb2f7f1.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "9ee6dd4f4a9b932d686c62a0fa553532fe37613b6e8adaa4c980692be5d8a71b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e0b795e7ea7f6b64e1ed8b2fbfd281f21a0443aaf223e457ab59d5f085b9d14d", "generation_seconds_batch": 22.725412467028946, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.38003396987915, "genuineness": 4.287685871124268, "identity": 0.5131804943084717, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9ee6dd4f4a9b932d686c62a0fa553532fe37613b6e8adaa4c980692be5d8a71b.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "9efcc9be383169fc5cd7b1dde50909e17af1ed23e7d6d8910d3eed0c91218dcc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "980bb17324025b7418c7eb36ab1cf6eef67bd853d075fcdff7a20a0bc7dd72fa", "generation_seconds_batch": 15.821651450009085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.284721851348877, "genuineness": 4.256305694580078, "identity": 0.567797064781189, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9efcc9be383169fc5cd7b1dde50909e17af1ed23e7d6d8910d3eed0c91218dcc.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "9efeadc340c65a0ec7924b9266b47a10c43865dc459cc1455857a236cd94ac7c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "019bbc20587f6064ee946c775b5855aaf0ea131d9b51b13d796f412c7081db8d", "generation_seconds_batch": 16.48630112200044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5744330883026123, "target_z": 1.679687509940761, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.5660060048103333, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9efeadc340c65a0ec7924b9266b47a10c43865dc459cc1455857a236cd94ac7c.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "9f2cebbd366f370aeac3f76e030bb0d68e410b6a44f030863c526a7344fe2e59", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df18b10f9ba73ad54541a46fe8cf4b286178e0e9fc42f17211a083111408f49f", "generation_seconds_batch": 23.663462784024887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.141753673553467, "genuineness": 4.3504462242126465, "identity": 0.7055950164794922, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken. En", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9f2cebbd366f370aeac3f76e030bb0d68e410b6a44f030863c526a7344fe2e59.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "9f882456703694d8a047c4e8cb3b916a2521a88a0cb343063ac3836eb5b879ea", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aee8836dafd436475d87032e50363d5dba192aa94fb0476b30877c37074e015d", "generation_seconds_batch": 16.72356389206834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.173524379730225, "genuineness": 3.8169827461242676, "identity": 0.22715061902999878, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9f882456703694d8a047c4e8cb3b916a2521a88a0cb343063ac3836eb5b879ea.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "9fc291c1ac0c779c52f8382669084d813417dc92ea5f2d45aeef0b4299e15e35", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c39690664cd3101e347374104d30e05b0064987314c3bd0d41e1caf4bdeb1b89", "generation_seconds_batch": 22.665749451029114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.1020402908325195, "genuineness": 4.375550270080566, "identity": 0.622150182723999, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/9fc291c1ac0c779c52f8382669084d813417dc92ea5f2d45aeef0b4299e15e35.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "a0145bd7beb5cbbcc968430fe7de82cd8fde2c2780fa5f6208b9ff19ac5b5160", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48ca198261595766bef7254df3b342ac1ddea633133a9473dd2374f9b865b331", "generation_seconds_batch": 22.930848905001767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 5.796247482299805, "genuineness": 3.986435890197754, "identity": 0.6208739876747131, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a0145bd7beb5cbbcc968430fe7de82cd8fde2c2780fa5f6208b9ff19ac5b5160.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a02adcd3c766bac8fbd18c10b35091521022161d6d5cc2748b3c5e5f6feb9f83", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "399d377ca74f0915592433de90a32e57aa7df935018b77a57c5107134314f564", "generation_seconds_batch": 16.851675446960144, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.356205940246582, "genuineness": 4.482243061065674, "identity": 0.4280497431755066, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a02adcd3c766bac8fbd18c10b35091521022161d6d5cc2748b3c5e5f6feb9f83.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a0369fba376f900c15960d30ee8b534dd9ff60eecd4a80566376445f9b7ab7a8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cf042d57911efc075ef154ce140ecb867247389c8c23975c07ae08c32fc64a3", "generation_seconds_batch": 17.40934676793404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7751637697219849, "target_z": 1.0234375374690228, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.4154980778694153, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a0369fba376f900c15960d30ee8b534dd9ff60eecd4a80566376445f9b7ab7a8.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "a16c26a295bd685200db95b13772114abaa6dd6fb205774e82d5d9f6154dc1f7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b919ed238159e8de63b56c9493b82644ee8e556efe01d5ba2e372af4780f06ab", "generation_seconds_batch": 25.043397782952525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.181467056274414, "genuineness": 4.168440818786621, "identity": 0.7669023275375366, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a16c26a295bd685200db95b13772114abaa6dd6fb205774e82d5d9f6154dc1f7.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a1e65b69568d6bbdb51fd1bcdce27f47bf54a784242e7e14eb6639491c858fd6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9256f723a58aa85bf55ae4e257d18d48e7dd363faad40865c1f88841d291dcba", "generation_seconds_batch": 20.822211446007714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.324435234069824, "genuineness": 4.319066047668457, "identity": 0.3675588369369507, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a1e65b69568d6bbdb51fd1bcdce27f47bf54a784242e7e14eb6639491c858fd6.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "a22a0a3203bb8afa4f4386fbc9b2f13a67f677b41a76c1e55a44c5aba766cc9d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bce653abb9f97d6bfcb1e3c1e85540e1686c9be95e8fdcfbfedbba6ec3ff1f9d", "generation_seconds_batch": 23.95085292297881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.33808034658432007, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a22a0a3203bb8afa4f4386fbc9b2f13a67f677b41a76c1e55a44c5aba766cc9d.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "a28649f24cbc7a081aa19be40749bbd2a54dabee628c7cc81fe72878b33f2977", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee2f926a48c8e9a89c42ae0363e36619bdf40106c0eca86b6da6e652de8eb0ef", "generation_seconds_batch": 22.53786337794736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 5.895530700683594, "genuineness": 4.519899368286133, "identity": 0.5924715399742126, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a28649f24cbc7a081aa19be40749bbd2a54dabee628c7cc81fe72878b33f2977.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "a2ea3add4389f4f2ac38e4c5845201a77c1294dc39e1f50100db7f60bc26465e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4e7ca93d9ce2e12c104419cf6d01825e0de68677473217de4134b6f57452ba9", "generation_seconds_batch": 24.189882898004726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.0146708488464355, "genuineness": 4.143336772918701, "identity": 0.8041294813156128, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a2ea3add4389f4f2ac38e4c5845201a77c1294dc39e1f50100db7f60bc26465e.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "a31019e7e5fafa56e9d5c3246565fee93ef1eeb79babd1c7d5d25fb53cd0c0d7", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51b5aea633115e97278fb8fefc3923102fa2cfe901aefb9dc7aa615e5ddda7df", "generation_seconds_batch": 16.06706341705285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 5.95112943649292, "genuineness": 4.011539936065674, "identity": 0.6985880136489868, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a31019e7e5fafa56e9d5c3246565fee93ef1eeb79babd1c7d5d25fb53cd0c0d7.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "a318aca24c30ad925b387853f883f9298c7ccca5e4cb6bca0c70b28104a829f3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de7898aedf2eedd2d4e762fb7b6e3f6ddab384432a47ceb46df74ac2f9907a70", "generation_seconds_batch": 23.140089018968865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.165581703186035, "genuineness": 3.9550557136535645, "identity": 0.6264362335205078, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster, wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a318aca24c30ad925b387853f883f9298c7ccca5e4cb6bca0c70b28104a829f3.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "a33c9e927bc3343100b3b9aaf9ffb9039ee57db165eca6c96253dbd4d68146cb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caefb45ce0b2d9f8077859c209193c2b4c90e2fe269fe85947769798739b241a", "generation_seconds_batch": 22.10166665702127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7076447010040283, "target_z": 1.7890625542918492, "ce": 6.181467056274414, "genuineness": 3.7291181087493896, "identity": 0.4238927364349365, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a33c9e927bc3343100b3b9aaf9ffb9039ee57db165eca6c96253dbd4d68146cb.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "a34647ca934885126c88317aaf88aa2cc0483afa9b6830c8475dc18cd91556c1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0c93aa1766b83a9de0c5200a8afb74fb4ec96669419c74b405f416bdc6e569ea", "generation_seconds_batch": 23.054461400024593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.6128666996955872, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a34647ca934885126c88317aaf88aa2cc0483afa9b6830c8475dc18cd91556c1.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "a417bd40c89dfa0f6524afbb9ea25413b9893cf4af099684c556f6d38a553bae", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7dbe5eb1f182782fce94c3da0ee2b0d22514975f88f416f3fdb5030df0fb3fd", "generation_seconds_batch": 23.802937470958568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.213237762451172, "genuineness": 3.8044304847717285, "identity": 0.18737304210662842, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a417bd40c89dfa0f6524afbb9ea25413b9893cf4af099684c556f6d38a553bae.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "a41ce0f9bdcea8d51bc6831ed6cc297d846e6c11c85bd60be449b5f6697cabc4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30abc93beda67aeccbe77a30d8f3e5764debae3f702a2b2d203562f54aa50131", "generation_seconds_batch": 16.889136526035145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.515059471130371, "genuineness": 4.231201171875, "identity": 0.601789116859436, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a41ce0f9bdcea8d51bc6831ed6cc297d846e6c11c85bd60be449b5f6697cabc4.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "a41dc58d988012564999d0881fc2578c65cd5a6f53dcd9669d0f49e9bb746406", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d770f36d4436ef9a3adc451a79868c6efaa0d8f9c9d7d8248743fc1d7c77486", "generation_seconds_batch": 23.023433946073055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.181467056274414, "genuineness": 3.7918784618377686, "identity": 0.41331666707992554, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a41dc58d988012564999d0881fc2578c65cd5a6f53dcd9669d0f49e9bb746406.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "a44991cfe238ef1985003caaf1734bdc6bc24432a5863223bca0f467eb44790c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1401049630951379e2c46f0017c39ef2c751de5d0d9b8622e49582e79aacbead", "generation_seconds_batch": 15.438471636036411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8512846231460571, "target_z": 1.0859375068820654, "ce": 6.252951145172119, "genuineness": 4.6705241203308105, "identity": 0.6426897048950195, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a44991cfe238ef1985003caaf1734bdc6bc24432a5863223bca0f467eb44790c.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "a44ca18859cf6f5b9ae513c64b4c9a1fee1ed02b4e5cf7285b7e66fc61dde7b2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9d82e87a29ced9c6a2095d752b6392914a9b17e881a600653bf4068a84501ba", "generation_seconds_batch": 22.920538848033175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 5.974957466125488, "genuineness": 4.369274139404297, "identity": 0.6426271200180054, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a44ca18859cf6f5b9ae513c64b4c9a1fee1ed02b4e5cf7285b7e66fc61dde7b2.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a4d9ef9dcfc80a38601080660172e668c3b89cf679e21c31c60acbaeeba7ff0a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f163a39c421c48e774383d2c05321ba5e088620468a05e874684c9d5d3f813c3", "generation_seconds_batch": 23.08624123397749, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8893450498580933, "target_z": 1.1171874915885867, "ce": 6.221180438995361, "genuineness": 4.149612903594971, "identity": 0.3908314108848572, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a4d9ef9dcfc80a38601080660172e668c3b89cf679e21c31c60acbaeeba7ff0a.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "a51ca26a626cb987fd48abbb0582ef1ed31c0e4837f41f8630544897842874c5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8379fc99457b041133e4e3fde171da4ce05939aa216e17751647c7238107f79e", "generation_seconds_batch": 23.474074197001755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.592550039291382, "target_z": 2.5156250412923926, "ce": 5.919358730316162, "genuineness": 3.7542223930358887, "identity": 0.7220312356948853, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a51ca26a626cb987fd48abbb0582ef1ed31c0e4837f41f8630544897842874c5.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "a5589d3032c1b7d877e6fba4efa27d6abe327240176e36eb205c28bfba88f1b0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55e8838464c430a6828e24b461b0904dcab0995d42afb6278701a7d8d7cb128a", "generation_seconds_batch": 21.83630969701335, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.181467056274414, "genuineness": 4.2751336097717285, "identity": 0.660487174987793, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a5589d3032c1b7d877e6fba4efa27d6abe327240176e36eb205c28bfba88f1b0.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "a560a16d7826406cae849b15bd2a309fa0c8933ffa10c4016d17607ef16011dd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4caa8385e0fdea44c0e6da8b60f9f2f599a35c8e3eaeb8f8b4ea2c109242d688", "generation_seconds_batch": 17.601493925903924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4221911430358887, "target_z": 1.5546873753581487, "ce": 6.332377910614014, "genuineness": 4.419482707977295, "identity": 0.7281798124313354, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a560a16d7826406cae849b15bd2a309fa0c8933ffa10c4016d17607ef16011dd.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "a5eda63f3eb6990a9eaed4f4d6c44a7f356ebbf33bd923cc177d7c81e3065581", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b49fa19c8d7e30f7171e4313b686a827a7341e59da1c8e3288cd5872d6e1c2a", "generation_seconds_batch": 23.13862421794329, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.7063677310943604, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a5eda63f3eb6990a9eaed4f4d6c44a7f356ebbf33bd923cc177d7c81e3065581.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "a63e2995a3a060be069a35233164627f1058b2d91bef06a740c77befff7914e7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c8c2942a7ea507edcdb551eb135f02f70a06a2b16ec41c20e082fafc7bea90a", "generation_seconds_batch": 23.341663270955905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.070269584655762, "genuineness": 4.20609712600708, "identity": 0.7255700826644897, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a63e2995a3a060be069a35233164627f1058b2d91bef06a740c77befff7914e7.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "a651634ca59c59a43c0ad8e6d4a576056767a232e3da3eccc056377bf0dccc0f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "787c8081aaf9ef6bb0b4835d5e994ef86f944a645db8145b755667c6b2ec37a7", "generation_seconds_batch": 25.077130577992648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.125868320465088, "genuineness": 4.2374773025512695, "identity": 0.7445377111434937, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a651634ca59c59a43c0ad8e6d4a576056767a232e3da3eccc056377bf0dccc0f.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "a6ec3e55335ccd6a042596bd3f3776ee3eae5fbcc32927a98744a58a2203f8f4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3995b5b131e92edc8d8f7d83240ab43fa0ee7bc8089e5fbd5fd5e6ad2f702fee", "generation_seconds_batch": 16.643371348036453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.4783790111541748, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a6ec3e55335ccd6a042596bd3f3776ee3eae5fbcc32927a98744a58a2203f8f4.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "a7eb026eb8f39d85dda07ef4e59a63c6a7d7502887dc1a8dbbc5caf75f6ad508", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8382aacf651d9e76baa497181e9f3d56f303226d3262ad6f92fddf6a1dcfd71b", "generation_seconds_batch": 25.077130577992648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.7105736136436462, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a7eb026eb8f39d85dda07ef4e59a63c6a7d7502887dc1a8dbbc5caf75f6ad508.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30581a59c6ba77495878419dd31e1b0a75d78980018d7918002db7780330cd6a", "generation_seconds_batch": 16.8535391719779, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.7091017961502075, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "a8d768225d18db38e6fbaa89cf13e842c9ba6c244c1dac42c0d71be97a65f56c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a82c9c1912cf2e6a60957d24a64c3288976c7b54c7cf74af5a35a45ba10e417", "generation_seconds_batch": 16.053499951958656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6115801334381104, "target_z": 2.5312499357673897, "ce": 6.4912309646606445, "genuineness": 4.005263805389404, "identity": 0.5047541260719299, "hypothesis": "De weg gaat linksaf na de brug. Loop verder, tot aan het kleine, houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a8d768225d18db38e6fbaa89cf13e842c9ba6c244c1dac42c0d71be97a65f56c.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28317c352215d543c9eb6e62a85b9bc14a64c8cbef4728adf660b5ad8d825851", "generation_seconds_batch": 16.9775777120376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.648396372795105, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "a94cab3681c852973930493d353c9b745055198a6c6d4a117648c80c5e8a6cb5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "982db81cad58b9cea8b4daa6928aa2f01b21d0b6742b6eceeda035a0924d4965", "generation_seconds_batch": 23.43190095096361, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 6.316492557525635, "genuineness": 4.099404335021973, "identity": 0.4155287742614746, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a94cab3681c852973930493d353c9b745055198a6c6d4a117648c80c5e8a6cb5.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "a983b9ed386c6d19a298552198c3b5d912f21694020bc7076fa014bf972332f7", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I wrote the number on the envelope.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cb5e9716b0edaafa46fa974297725ab2d7b2709aa829b232c416f48273d14e6", "generation_seconds_batch": 16.78895953495521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.435632705688477, "genuineness": 3.7793264389038086, "identity": 0.4663366675376892, "hypothesis": "I wrote the number on the envelope. Read it carefully and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a983b9ed386c6d19a298552198c3b5d912f21694020bc7076fa014bf972332f7.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a99d50bb7f265e881a9bcd3c8a17496b29f7b8d49129fedc1639535d50469794", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bf1a176b34d8b9f550d2e76d3f56a7e73dd6482ead75fe757091f26cc7e9a92", "generation_seconds_batch": 17.91126184805762, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.292664527893066, "genuineness": 4.1998209953308105, "identity": 0.28423944115638733, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a99d50bb7f265e881a9bcd3c8a17496b29f7b8d49129fedc1639535d50469794.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a9d18e54eea4ed8379e0639439c1c10430383ee65521e542789874ca4e695d98", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1e0f02702b44f627847902aac83743b7372071d1e9f4861b36cee711794a22b", "generation_seconds_batch": 16.63660972798243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3841309547424316, "target_z": 1.5234375864081544, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.6552317142486572, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler ?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a9d18e54eea4ed8379e0639439c1c10430383ee65521e542789874ca4e695d98.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "a9d1c07d5a4d1224272ed52b8335b689897fca9cfc3210355ff0e974040df924", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03f1adecc0e60221617864f9c8f87a2d59b070820c1203a6df60f881f68cd1a3", "generation_seconds_batch": 15.348337210947648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.060616970062256, "target_z": 1.2578124227679326, "ce": 6.372091293334961, "genuineness": 4.482243061065674, "identity": 0.35243552923202515, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a9d1c07d5a4d1224272ed52b8335b689897fca9cfc3210355ff0e974040df924.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "a9d28702e00ee4c5ba82598c4d8a341f3f65f9fa734f02d0560cfa81f5193966", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bc6f334c1b4ba64ae3928d08cf1ae14cf55ed07371c0a782c919edf05465dfd", "generation_seconds_batch": 18.712021928979084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.7406325340270996, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a9d28702e00ee4c5ba82598c4d8a341f3f65f9fa734f02d0560cfa81f5193966.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "a9e5015630e9a855c76cbf6fa20631c3b9b44ca560144ff0647b6c159db800e4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30c87830f170c3a629565f64f333913651565ab17ade446f578ff9f3af8b366b", "generation_seconds_batch": 22.712754824082367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 6.157639026641846, "genuineness": 3.9927117824554443, "identity": 0.41272252798080444, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/a9e5015630e9a855c76cbf6fa20631c3b9b44ca560144ff0647b6c159db800e4.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "aa23f3c9766de1ce8c9796d74751595c4a82c475af4d5c480e5645420ca13df8", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4ff96977827529095ceef8700b77e9cbc5e8b4f30fe8da0b1f56cfc11c4dab1e", "generation_seconds_batch": 15.98727486806456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 5.935244083404541, "genuineness": 4.143336772918701, "identity": 0.6501070261001587, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aa23f3c9766de1ce8c9796d74751595c4a82c475af4d5c480e5645420ca13df8.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "aa74dafeefbd86a90ce837a3bec2d39ee9e3d877309f43763cfceac536b53954", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "faf2af93d40b2322877a1127fc8dca08af5110867678326261f9be6d359b8411", "generation_seconds_batch": 23.121323812985793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1177077293395996, "target_z": 1.3046874977059781, "ce": 5.8875885009765625, "genuineness": 3.9927117824554443, "identity": 0.6161139011383057, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay después.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aa74dafeefbd86a90ce837a3bec2d39ee9e3d877309f43763cfceac536b53954.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "aa874e998d2d28bf6e05bc762e3b6f7da01884ae268d240d8137b6f29ff648ed", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ae1a99f16e43555c3e35a15d24d17b2f9519ae728e0aa96c6b6def181ee42f5", "generation_seconds_batch": 26.68256098695565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.070269584655762, "genuineness": 4.231201171875, "identity": 0.5432888865470886, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé, dans le tiroir, puis près de la fenêtre. Prends-les, quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aa874e998d2d28bf6e05bc762e3b6f7da01884ae268d240d8137b6f29ff648ed.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "aaacd257a93832687c9bad2716968bd9f9979423e44ad3d3cc75dbe90bf8487b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8af6206e04f31d3d915d81d0ce43710b5b531781b8e5f0d7e5d64afcd008a4c6", "generation_seconds_batch": 14.200745025998913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3841309547424316, "target_z": 1.5234375864081544, "ce": 6.24500846862793, "genuineness": 4.344170093536377, "identity": 0.6346924901008606, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aaacd257a93832687c9bad2716968bd9f9979423e44ad3d3cc75dbe90bf8487b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "aacb2fd9b843e914ebe8a07082d20aee9642b06844c0d9a0a63f37b74a46b996", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c314d4c646e3f67f5a4e01cce5232423f6f7d76ab1af2cf71cf64ffb28a8770f", "generation_seconds_batch": 21.652251077000983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.7153714895248413, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aacb2fd9b843e914ebe8a07082d20aee9642b06844c0d9a0a63f37b74a46b996.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "aaf568b2b603ce25a169c1be1ef8ab35c4f3e5dda004997acb29ab52629db645", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0390c656debaf1bfdac547e89757f7e6335adc6e2e76db47e4a09d865b6274d4", "generation_seconds_batch": 17.472881059045903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.165581703186035, "genuineness": 4.582659721374512, "identity": 0.771793782711029, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/aaf568b2b603ce25a169c1be1ef8ab35c4f3e5dda004997acb29ab52629db645.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "ab3e1b74f1dd6811efec58de5fa8187871f7e21a8fb9d4a9eb31af253d7d99fc", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0f147b4fc00ccf30b5108ddd67e6e01b753f62184c15b339288c547ac393a51", "generation_seconds_batch": 15.348337210947648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.300607204437256, "genuineness": 4.293961524963379, "identity": 0.4608210623264313, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab3e1b74f1dd6811efec58de5fa8187871f7e21a8fb9d4a9eb31af253d7d99fc.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "ab3f3f4d776fac82a09eef956e175bd7a2f891d32a714ef9c1615e5750ea15a5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7b1264c4afd6c2ac5b839c596dbbf83f9ffa5553a1b2866838c7934dcd0de05", "generation_seconds_batch": 23.022105142008513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 5.7684478759765625, "genuineness": 3.8671910762786865, "identity": 0.6922343969345093, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab3f3f4d776fac82a09eef956e175bd7a2f891d32a714ef9c1615e5750ea15a5.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06505f516d86bec160ee87441face2d7568a06bcef026028380f76cbcbd2e07e", "generation_seconds_batch": 16.994158094050363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5753463506698608, "target_z": 0.8593749709423906, "ce": 6.4912309646606445, "genuineness": 4.105680465698242, "identity": 0.170420303940773, "hypothesis": "Your message arrived after I had left. I read it twice, before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "ab8098cbb12a61d070ee919f734c6fab70899133c77f67ec43d987c937016351", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f3d86e61ab86388d5d8ca3590ba4d6bd7bb319b0110449ddfb67bdd6b3c7b42c", "generation_seconds_batch": 23.535628710058518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.23706579208374, "genuineness": 4.375550270080566, "identity": 0.49810412526130676, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab8098cbb12a61d070ee919f734c6fab70899133c77f67ec43d987c937016351.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ab8a11c25871e38cd8e2a88a1bb57034a9670833c1a6b3b1c58cfff02307c4c1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2500bc46d9b3b52a6158839988efb9984fecbddb02a6f4a9db85f0764e856cc0", "generation_seconds_batch": 23.173846847959794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.149696350097656, "genuineness": 3.7291181087493896, "identity": 0.3828775882720947, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab8a11c25871e38cd8e2a88a1bb57034a9670833c1a6b3b1c58cfff02307c4c1.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ab8d8fde9a790041db0ef6513b55c487b347b417325a4b6bc1967175942ecce9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20a05dba8e0e4148c4f959170f41d5389e4742fb9a16aaa93bd47991ac2882ea", "generation_seconds_batch": 16.803584781941026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.23706579208374, "genuineness": 3.6412534713745117, "identity": 0.3537611961364746, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ab8d8fde9a790041db0ef6513b55c487b347b417325a4b6bc1967175942ecce9.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "abf827cdd442a864b97b75e7bc15a0ce7f482d0eb217ec842d8f95163fc4bf22", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d747cc9f943837df7639d14c4e2086c2f9d9c8249e8a4ff02e94d973e9250c47", "generation_seconds_batch": 17.050725597073324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.536372661590576, "target_z": 1.6484375252342398, "ce": 6.213237762451172, "genuineness": 4.293961524963379, "identity": 0.7448785305023193, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/abf827cdd442a864b97b75e7bc15a0ce7f482d0eb217ec842d8f95163fc4bf22.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "ac14c931da238f1d88863155bed6f9cd6dc00d82fb74ba6f5b26cb37772908cb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31d66a5b69e5f780aa4ed6773bdd3ae808714f85e42ae7ecb93b28e9ecd8c8f8", "generation_seconds_batch": 15.39561661996413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.483288764953613, "genuineness": 4.13078498840332, "identity": 0.4514676034450531, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ac14c931da238f1d88863155bed6f9cd6dc00d82fb74ba6f5b26cb37772908cb.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "ac7c843e6bbc2d226830b37656bd5768302579e79d6a0164cdd2ab48d44bde52", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5f13a2f28feed26bd611ad612b25dc4ce03445ab1c38d4588ab6fe433a76641", "generation_seconds_batch": 19.223884980892763, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.252951145172119, "genuineness": 4.1998209953308105, "identity": 0.23254798352718353, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ac7c843e6bbc2d226830b37656bd5768302579e79d6a0164cdd2ab48d44bde52.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "acc677e9c557d25393180776f7743848eab3d5288ce1351a7c262fd9e166dec1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "24c7a92ae91f1325ff2d7f8b3081eedb6b39c94f8a1493525e62418edde46760", "generation_seconds_batch": 16.15053148206789, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.181467056274414, "genuineness": 4.18099308013916, "identity": 0.7444442510604858, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/acc677e9c557d25393180776f7743848eab3d5288ce1351a7c262fd9e166dec1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "ace13466bf2f8c03b7104da3f4778ad7452a56d1159cd27741ace4c48fe56380", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b7cdce152d61eebc3492d243bf5fa043b91e845cecb1c8ed1d1944807692972", "generation_seconds_batch": 17.544126479071565, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.40316104888916, "target_z": 1.5390624808831517, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.7286974787712097, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ace13466bf2f8c03b7104da3f4778ad7452a56d1159cd27741ace4c48fe56380.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "acf682e918845c5869d0c17e8a605fb19004bbf662260cf9e54ea996915d1279", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a4ef6137693d50e3f13df02f4ab9124a6ddc0efcbad2c6e2ed353b9ca541b06", "generation_seconds_batch": 22.842010893044062, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3451571464538574, "target_z": 2.3125000428217404, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.5184141397476196, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/acf682e918845c5869d0c17e8a605fb19004bbf662260cf9e54ea996915d1279.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "ad0dfb8958e1d5bc5142817980ba1db47b4cfe2a6a60d073e69fdc0554d425a7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d224fca76c76d07744ded7c66eb5d9e62b1d1b9eafa1812ef2b57f09e5e97c0", "generation_seconds_batch": 24.574281575973146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9369206428527832, "target_z": 1.1562500214108702, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.3245176672935486, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad0dfb8958e1d5bc5142817980ba1db47b4cfe2a6a60d073e69fdc0554d425a7.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ad2faf19b3a9f616b2b207213ac8f9079ba40a453d804b3d507eebec9407c4f4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b3d0c40dc211feb5547fa43f3d9df3ad8e0ff439a64788a484d86aa11ee6e0c", "generation_seconds_batch": 17.125282229972072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.197352409362793, "genuineness": 3.7793264389038086, "identity": 0.49677374958992004, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad2faf19b3a9f616b2b207213ac8f9079ba40a453d804b3d507eebec9407c4f4.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "ad5420cfa7c0ee94dfd52511585e3cd9114ad9ce391b1574c340f8d2067fdc92", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bae809a5372ca91e839c9084ec2b32422f7c33843c16c5d89572bb726458189", "generation_seconds_batch": 24.355041628936306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.372091293334961, "genuineness": 4.111956596374512, "identity": 0.42724722623825073, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad5420cfa7c0ee94dfd52511585e3cd9114ad9ce391b1574c340f8d2067fdc92.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "ad54252615000349f07dee3555538980c2c976798fd184d7c358d187f0f5f650", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "680db9fee995a54d3adc4aa68065875f3a3cb02542b289c2a1c08339cc3eb8e7", "generation_seconds_batch": 24.52314870490227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.660982370376587, "target_z": 0.9296874854711953, "ce": 5.601652145385742, "genuineness": 4.432034492492676, "identity": 0.8149996995925903, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad54252615000349f07dee3555538980c2c976798fd184d7c358d187f0f5f650.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "ad68b9f560fc06b327541608f7598bc8e4bc8033a2046eac3f56bab222c1c82f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91d4506a8faddc2c92ae67db49c61fa9ddf09b5a8a7a2afbb7846c0d1841a18d", "generation_seconds_batch": 17.63052460399922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2414040565490723, "target_z": 1.4062498990630408, "ce": 6.300607204437256, "genuineness": 4.406930446624756, "identity": 0.5477822422981262, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist. Berrie.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad68b9f560fc06b327541608f7598bc8e4bc8033a2046eac3f56bab222c1c82f.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "ad719ed76826a11bf3497c945f1315909df7ee0ec9de4d0b78af34e6a1afd9fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ad8387fc6d665c2e072ce4a5dec4bc7f6894d37b8d7466803eee5c9873e57650", "generation_seconds_batch": 22.570028182002716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7561335563659668, "target_z": 1.007812545115762, "ce": 6.181467056274414, "genuineness": 4.7458367347717285, "identity": 0.8304013013839722, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad719ed76826a11bf3497c945f1315909df7ee0ec9de4d0b78af34e6a1afd9fa.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ad7cdb77dd22c5422299e6e87743ebefad48d507d115e162c3e02ab02950d91f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d45354543acc5ff806c2b565cb158273812581ecaa485b9f7306975b0329a09d", "generation_seconds_batch": 17.063161210040562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.284721851348877, "genuineness": 4.218649387359619, "identity": 0.2624402344226837, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ad7cdb77dd22c5422299e6e87743ebefad48d507d115e162c3e02ab02950d91f.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "adbcc73ffebb54103e65821c23d7857e817006c5fbe9957ca40de70ff187a42c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ce3bd041a91330d554585c72479ceba6b3c1c9a2fc19a17f0e7702fa5e3e74f0", "generation_seconds_batch": 15.821651450009085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.38003396987915, "genuineness": 4.325342178344727, "identity": 0.5077280402183533, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/adbcc73ffebb54103e65821c23d7857e817006c5fbe9957ca40de70ff187a42c.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "adc887e76a350dfd2ec90d2e5d7e5d32efe89f1c2a27bb1551fd94a8bd48e29a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b26f5d3f77ab717f11c8a646163b346365cd9f21b4ac8e00bed05aba722d2e9b", "generation_seconds_batch": 21.44706320797559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0415868759155273, "target_z": 1.2421875282929355, "ce": 6.284721851348877, "genuineness": 3.998987913131714, "identity": 0.4620702266693115, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/adc887e76a350dfd2ec90d2e5d7e5d32efe89f1c2a27bb1551fd94a8bd48e29a.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "295208bc75c438128efc6bb3565c8cc66b601c8b50fa3159734536e06b7da64b", "generation_seconds_batch": 16.99337484803982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5230133533477783, "target_z": 0.8164063153796214, "ce": 6.070269584655762, "genuineness": 4.055472373962402, "identity": 0.6978772282600403, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "addc22f7d91072b0eaa16e06960d3a0de517398f7482ca6b439fd0b50343d9fb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "77efd7f8a3f2c2a5b9efae6c1db145f63a5276b1134c6633d890e0bd763796e0", "generation_seconds_batch": 21.97664772195276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460251808166504, "target_z": 1.5859375558211972, "ce": 6.395919322967529, "genuineness": 4.105680465698242, "identity": 0.5800157785415649, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/addc22f7d91072b0eaa16e06960d3a0de517398f7482ca6b439fd0b50343d9fb.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "ae191e603c1a0c8906f9315a1736e21d5a4db529448a195e71683f922a3839b2", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca359b973b5746a640f62dede4d6252b36793554e598e0ef6933d929e133119c", "generation_seconds_batch": 16.866452172980644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.7185051441192627, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ae191e603c1a0c8906f9315a1736e21d5a4db529448a195e71683f922a3839b2.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The road turns left behind the bridge.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15c90145f9949ce7d31d1fa00c362e85937ab1fca15d151451b6b47dbeef7392", "generation_seconds_batch": 17.281980382977054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.38797664642334, "genuineness": 4.005263805389404, "identity": 0.1867043673992157, "hypothesis": "The road turns left behind the bridge. Keep going. Until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ae4190335c925e96dfad8b9b0c55665db0d1848a25c54413d49dc976d7e2b7ff", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7c939a4f3e27f800830852929ee54ad0f8d740c1296774e9c786e84ab2432ae", "generation_seconds_batch": 23.65285486797802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.221180438995361, "genuineness": 3.760498285293579, "identity": 0.42218390107154846, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ae4190335c925e96dfad8b9b0c55665db0d1848a25c54413d49dc976d7e2b7ff.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "ae6a8bf7b5b4143a540c0f9347dc1fb34ab6358d1943d00ab847a61173508cdb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8128ddb13976dfe9583cebff5e95a575c6d50e94861e12d3032dd3109d2b32af", "generation_seconds_batch": 22.968566749012098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.292664527893066, "genuineness": 4.369274139404297, "identity": 0.39744192361831665, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ae6a8bf7b5b4143a540c0f9347dc1fb34ab6358d1943d00ab847a61173508cdb.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "af83f98a2fdf76667b522d52914c16d6133e90a0c1b9bcabc800dfde47d9fb2f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56f6e7f69061c9de3d0db9f05976673eec5052fcd170aa1e20180d6fb174c3e6", "generation_seconds_batch": 24.12653525196947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3832175731658936, "target_z": 2.3437500275282614, "ce": 6.268836498260498, "genuineness": 4.118232727050781, "identity": 0.6748306751251221, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/af83f98a2fdf76667b522d52914c16d6133e90a0c1b9bcabc800dfde47d9fb2f.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "af9ab98ef2682d0d5b0f410bedb722bf9a4ea3dac45e3ae7c527d6c2cbb6605b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b579ecb5576c45c574088e3b7764ac0494cdf98d0ffb143b8632782b47376905", "generation_seconds_batch": 15.109691446996294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 6.308549880981445, "genuineness": 4.325342178344727, "identity": 0.21922056376934052, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/af9ab98ef2682d0d5b0f410bedb722bf9a4ea3dac45e3ae7c527d6c2cbb6605b.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "b1f5ebdc2c91f654728dba10192da239255622d828ab658cef5eb1779dd836f2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "99c8677a1d7907352f81f85e183930270197d433919554973620b72811ddffa1", "generation_seconds_batch": 16.904202364035882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.038498878479004, "genuineness": 3.8797430992126465, "identity": 0.7755212187767029, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b1f5ebdc2c91f654728dba10192da239255622d828ab658cef5eb1779dd836f2.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b2043334ccf7e7921c71b61607a4578b98cf11716fcde25a0a80003621102a67", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "089fdca53f255d092c0f1e8e64234dc90761e3b8f5d73f403c2f8d3428ad3b90", "generation_seconds_batch": 21.94106747605838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.165581703186035, "genuineness": 3.7667744159698486, "identity": 0.39244428277015686, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b2043334ccf7e7921c71b61607a4578b98cf11716fcde25a0a80003621102a67.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "b25e228258ae4757e1cf2f2d0ab34acb5285f1f20d832b5478d06f082aeabb01", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e89af7aef1699a4984ff61453379955659646c43a557c1ae23a18efd66ecc459", "generation_seconds_batch": 27.696059879031964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 6.125868320465088, "genuineness": 3.998987913131714, "identity": 0.5257418155670166, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b25e228258ae4757e1cf2f2d0ab34acb5285f1f20d832b5478d06f082aeabb01.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "b27bb98db66a8f355f5a159c7daf5de50dc0ea5d26d268de3f1cff0c810a3051", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8b643431dcce529f7d3acb419ffded0c4080a7caead16fae32f472786764f0e", "generation_seconds_batch": 14.655249343952164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.1894097328186035, "genuineness": 4.231201171875, "identity": 0.8642802238464355, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b27bb98db66a8f355f5a159c7daf5de50dc0ea5d26d268de3f1cff0c810a3051.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "b33014eaf3c0354eb8c8cbcf064f61339c3269f24e3a2bf85665dde4ffca833b", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f4384cd002c2d5279f6ce455f9c7e877008cd77590b0dc4f90d828d12258f17", "generation_seconds_batch": 23.42861358600203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.141753673553467, "genuineness": 4.632867813110352, "identity": 0.7019796371459961, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b33014eaf3c0354eb8c8cbcf064f61339c3269f24e3a2bf85665dde4ffca833b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "b3c8f926fbbd7a3222fb33048c295e8bf4deea1ca5dbfea7e6e3e33f8a10cf2c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f64ee31460239870fe2d74f6591b2c35265572ccda8f5348bdcbccce22467c06", "generation_seconds_batch": 15.832985684974119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.141753673553467, "genuineness": 4.005263805389404, "identity": 0.6454620361328125, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b3c8f926fbbd7a3222fb33048c295e8bf4deea1ca5dbfea7e6e3e33f8a10cf2c.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "b3c910a171ea98069277d4b5164a36bec444a98504efedf21d88b76967a38605", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab9f230bc6e1b50a946e0b55e31cccb51fc8d82fe054dbb381e36e9d99750451", "generation_seconds_batch": 22.665749451029114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.6148345470428467, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b3c910a171ea98069277d4b5164a36bec444a98504efedf21d88b76967a38605.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "b3f4d1e160fec2b36c64d6758535e1e715d136c6b1d2b56c08a442caeb7328f2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32a99cbcc66b867534264ac91bc6d71ca8b2ddd4b58771caa89ae33f6dcfd67a", "generation_seconds_batch": 22.11672960105352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.38797664642334, "genuineness": 4.413206577301025, "identity": 0.6000440716743469, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b3f4d1e160fec2b36c64d6758535e1e715d136c6b1d2b56c08a442caeb7328f2.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "b40b356f7e9419c3c4587da40e2a700237d85872551ac7b726dea24ba2bdffc5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cc6f009e0254471f167d3f5f0d4774cac3f5b17680201764139fd7bc83d2714", "generation_seconds_batch": 22.64533573400695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.165581703186035, "genuineness": 4.2751336097717285, "identity": 0.4975478947162628, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b40b356f7e9419c3c4587da40e2a700237d85872551ac7b726dea24ba2bdffc5.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5d9297a337bb35cd9b1f86be05d0f6e16ed06fc496e0b3a0ed232af64c631ca", "generation_seconds_batch": 16.753720266977325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.435632705688477, "genuineness": 4.331617832183838, "identity": 0.5412794947624207, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b46da1c5d4fc781f02eb394f07417f0f9551af6ddcea4d9f3c7c8d9c3230c8b4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4fbb0fa72d9569aa45e477472d5334b6b8b34aa6208f6cef76cdf8e9c58cacc", "generation_seconds_batch": 17.886718077003025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.141753673553467, "genuineness": 3.7793264389038086, "identity": 0.23871582746505737, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b46da1c5d4fc781f02eb394f07417f0f9551af6ddcea4d9f3c7c8d9c3230c8b4.mp3" }, { "id": "heldout-nl-4", "language": "nl", "sentences": [ "Ik schreef het nummer op de envelop.", "Lees het goed... en controleer de laatste twee cijfers.", "Bel me wanneer je even tijd hebt." ], "text": "Ik schreef het nummer op de envelop. Lees het goed... en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-4", "key": "b50f0c951843bda929e3cdfa4746af2f4f62cd404b23a54c6a257fffc61da63a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d6afcaf938605e4617cc7b0067049efb623620b749db35af705b8a3ddfe0a5c3", "generation_seconds_batch": 23.090415695915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5544896125793457, "target_z": 2.484375056585871, "ce": 6.1894097328186035, "genuineness": 4.168440818786621, "identity": 0.4971955418586731, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b50f0c951843bda929e3cdfa4746af2f4f62cd404b23a54c6a257fffc61da63a.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "b51f30e2ccc24a6d122ff16742b75d877c3751014a8e58feb9b5444802682ead", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4774176e70fbbda89699c8dabbcba25e247eeada5d831b5f17c2137502df21c0", "generation_seconds_batch": 25.967078002053313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.372091293334961, "genuineness": 4.293961524963379, "identity": 0.723200261592865, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg: zeg maar, wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b51f30e2ccc24a6d122ff16742b75d877c3751014a8e58feb9b5444802682ead.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "b5a17972d5f02fcc3514f29a88224b597b215d7fcd24c06e4f53994191383dcd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81428b929df536aeebcb1686270a9a465d3f3e99d7e07b05b00875c4c26dd4a9", "generation_seconds_batch": 25.816367791965604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.213237762451172, "genuineness": 4.482243061065674, "identity": 0.7064427733421326, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b5a17972d5f02fcc3514f29a88224b597b215d7fcd24c06e4f53994191383dcd.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "b5c05f3156f3d6b944de1a1e8b6db938d7d211452b66c8041a0a548fad634e50", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aad9486d0e4e40107fff38c5a0398d12c10b0e2e72b5c811acbf6e83b14503b5", "generation_seconds_batch": 16.66987349302508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.7482353448867798, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b5c05f3156f3d6b944de1a1e8b6db938d7d211452b66c8041a0a548fad634e50.mp3" }, { "id": "heldout-en-2", "language": "en", "sentences": [ "The road turns left behind the bridge.", "Keep going... until you reach the small wooden gate.", "I will meet you on the other side." ], "text": "The road turns left behind the bridge. Keep going... until you reach the small wooden gate. I will meet you on the other side.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-2", "key": "b5e43d6dc2d3157fa20bea1c72c9efa45fb31b71e8ba379e3df00ece5328bdf8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The road turns left behind the bridge.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33922dfd128ec951af116019419107f402360db2834ca45a9ce92b00c75afd93", "generation_seconds_batch": 20.91312189400196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.403861999511719, "genuineness": 3.9425034523010254, "identity": 0.381076455116272, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. I will meet you on the other side.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b5e43d6dc2d3157fa20bea1c72c9efa45fb31b71e8ba379e3df00ece5328bdf8.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "b699169fcb2e0c871f6cb47d46f52b5724f16d4f9abe7558883f9d8f7e938579", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "206fcf292fe56ee83799c42ca91d4b6b459240988f221a35a1fbb4538f0fdc72", "generation_seconds_batch": 23.345343463937752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.7833314538002014, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b699169fcb2e0c871f6cb47d46f52b5724f16d4f9abe7558883f9d8f7e938579.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "b6abd36fcb9d188b20773e6e1c4bdb0e753420ba12145658d5b53bfff598af32", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b636d771ec56f0cf5fa6ffe5d16924e8d808136f34c2dff8fad66722214d8d5", "generation_seconds_batch": 18.662174213095568, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9369206428527832, "target_z": 1.1562500214108702, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.7232667207717896, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b6abd36fcb9d188b20773e6e1c4bdb0e753420ba12145658d5b53bfff598af32.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "b6e59e1763122f28e7f4a518bf73701678d490bff41038b361467d79f3ec10ed", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09941c0dc0d5afe63d1eacd7336cdc73fe5b2994198f39782209dada105caee8", "generation_seconds_batch": 22.340496483957395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.173524379730225, "genuineness": 4.419482707977295, "identity": 0.4172500967979431, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b6e59e1763122f28e7f4a518bf73701678d490bff41038b361467d79f3ec10ed.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b6f36ee76c0f084054472e44d7f98cfbade0e8e2063ebcd2ea1ef10da0b93cc1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3dc64f405d982887e53e927dc5c3d1c64811f34018f50b96dfc74cde14067235", "generation_seconds_batch": 16.286382565973327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.213237762451172, "genuineness": 3.8922953605651855, "identity": 0.18894121050834656, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b6f36ee76c0f084054472e44d7f98cfbade0e8e2063ebcd2ea1ef10da0b93cc1.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b6ffb90c7f6dfdcf14674f41eac848170ab9b682ce699a690cc830ab2bd21e0e", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0ae9d19f4806560a7c07546406f5bf813ac47316fbf7b306ef040edafced4cf", "generation_seconds_batch": 16.549172737053595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.205295085906982, "genuineness": 4.193545341491699, "identity": 0.48391371965408325, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b6ffb90c7f6dfdcf14674f41eac848170ab9b682ce699a690cc830ab2bd21e0e.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "b7104722e86da510b1767b8a05c56d267c20adf231e6447675f8968ea87da022", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca23ca7b413e1cbf254b321adc6bedc9620d4f860529a711f556ff041a756225", "generation_seconds_batch": 22.51104962802492, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.157639026641846, "genuineness": 4.13078498840332, "identity": 0.6411488056182861, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b7104722e86da510b1767b8a05c56d267c20adf231e6447675f8968ea87da022.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "b740eeda5347a1f3e520a821a6ac5caa7c08c498d9fe1e6e907d5250fa3d7fdc", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11bd0c4dab9612b1febd5589ed9ca4fa51dfa0a57376ee3dec637702bf02cfea", "generation_seconds_batch": 25.678539952030405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.098677635192871, "target_z": 1.289062603230981, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.6744676232337952, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b740eeda5347a1f3e520a821a6ac5caa7c08c498d9fe1e6e907d5250fa3d7fdc.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "b7ad3050a39d7abf8d0a4e446710701b35ed43c1979f44c4468b38bed1004797", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a28bbbd12bd8253058738f8cc99affc4a3e57ba8323460acfe5c0d62535c1768", "generation_seconds_batch": 27.39961539604701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.340320587158203, "genuineness": 4.570107460021973, "identity": 0.5065363645553589, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b7ad3050a39d7abf8d0a4e446710701b35ed43c1979f44c4468b38bed1004797.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "b8717c566a8ea04cd288eecbb5545a527fa4a58ba77953712eb7fd80ac841ea4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f705ac432473978903d863cf879f083b15f5376fb5316720e16bb40979f19d17", "generation_seconds_batch": 25.949483425938524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269036293029785, "target_z": 2.250000073408698, "ce": 6.268836498260498, "genuineness": 4.400654315948486, "identity": 0.46406322717666626, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b8717c566a8ea04cd288eecbb5545a527fa4a58ba77953712eb7fd80ac841ea4.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b8bb09b68600055329a13b8bd8bd4bc5e86501d89d3326f972a6bc5eb9a29668", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26cdc054593174610005a0e9f8c6e91d3676953a458cc5dbda47bb499364911a", "generation_seconds_batch": 14.86330408800859, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.149696350097656, "genuineness": 3.9676074981689453, "identity": 0.23256108164787292, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b8bb09b68600055329a13b8bd8bd4bc5e86501d89d3326f972a6bc5eb9a29668.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "b8dd1018c179f085936c53da44c3df49df7961e094400ac6c0fd21bf9c7d035b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf23f422a55c3780c37f8d37bb54ad43c23c13ab508ac8a654db346964e4bfb8", "generation_seconds_batch": 23.07975836796686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.7238792181015015, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b8dd1018c179f085936c53da44c3df49df7961e094400ac6c0fd21bf9c7d035b.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "b8fe0ab6599af3da32fe378eacc873c766a7043957d292a9be4f02561e71be42", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2edd68b426f4e461b363ac9781b16144f3fe87787be22b1a33d62a903df33b8", "generation_seconds_batch": 23.581734286039136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.09409761428833, "genuineness": 4.017816066741943, "identity": 0.5716636180877686, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b8fe0ab6599af3da32fe378eacc873c766a7043957d292a9be4f02561e71be42.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "b9010514831842157db9883e3c5c857f55eb01a7f7bac126d7621e1111520095", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ab3d9ca1160b108ec39f58f90beefb4544d389c11cd2e39b8e32c8dd3572eb1", "generation_seconds_batch": 16.547664126963355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.109982967376709, "genuineness": 4.105680465698242, "identity": 0.5234171748161316, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b9010514831842157db9883e3c5c857f55eb01a7f7bac126d7621e1111520095.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "b92f2e76de5b19eb0d47f92964320033396fe1f04e0d3da02f07c5491600f0a1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "24c1088c3669a0a653c6e4501ec2330ae8044757a26c4f71909b9dcb3769b7e3", "generation_seconds_batch": 23.573315830086358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.205295085906982, "genuineness": 3.9048471450805664, "identity": 0.4244765043258667, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b92f2e76de5b19eb0d47f92964320033396fe1f04e0d3da02f07c5491600f0a1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "b9687a20d6c47a9b5660e532dbd518566b7b11b167e0d5fec5427e1e672f9b4b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a79ec9f17050d06063e765f6ce998ef5ca7070d62dc4b0a8c38347f0490cf355", "generation_seconds_batch": 16.43211724993307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.6197206974029541, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b9687a20d6c47a9b5660e532dbd518566b7b11b167e0d5fec5427e1e672f9b4b.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "b96fa74c7f817edb159b806636c9c6cec0716481471eb1f94feb9598dc6a24cc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bcfb9515ff31163d251c14f62235b077ff796a0f0d6cf491213a04f4e4b2ba7c", "generation_seconds_batch": 23.84415880695451, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.3641486167907715, "genuineness": 4.21237325668335, "identity": 0.5854201316833496, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/b96fa74c7f817edb159b806636c9c6cec0716481471eb1f94feb9598dc6a24cc.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "ba09af3fb25f984d245e445cd5e1e18998feb4f60ec8342b232c3689af3b9ae1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1320d05fe6b47742e7f7b29f96fa7efed412a07388a8d858fadb10461e629f6c", "generation_seconds_batch": 27.696059879031964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8218259811401367, "target_z": 1.8828125084114133, "ce": 6.181467056274414, "genuineness": 3.923675537109375, "identity": 0.7672045230865479, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ba09af3fb25f984d245e445cd5e1e18998feb4f60ec8342b232c3689af3b9ae1.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "ba30807e2b131121590e12cd51ce46ac11cd64d28fe3b6bd6559f44a9dfbdfc0", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d400bf10eb000f1ac42c2e03f5b8a6fada0b9cfedb240f945ef459d4046ad504", "generation_seconds_batch": 17.463447876041755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.070269584655762, "genuineness": 4.356721878051758, "identity": 0.753455638885498, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ba30807e2b131121590e12cd51ce46ac11cd64d28fe3b6bd6559f44a9dfbdfc0.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a03ac5132fe7d804ddbc0b3d5a09ad275df939849dd2b48c2a6133d0de70978", "generation_seconds_batch": 16.755356175010093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.078212261199951, "genuineness": 4.369274139404297, "identity": 0.6254082918167114, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "bacfd4c918c8ffdc8dfed602436418d8faa75d59d579aea923819dfee8317088", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c52d884c6a8d265e67c454b905996125252a2c2639998f3171a6b6ba482127ba", "generation_seconds_batch": 16.028332632035017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3460702896118164, "target_z": 1.4921874059451061, "ce": 6.006728172302246, "genuineness": 4.375550270080566, "identity": 0.6184405088424683, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bacfd4c918c8ffdc8dfed602436418d8faa75d59d579aea923819dfee8317088.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "bb1a74d8d7680586ccaae79776b3abed91a4d02565ccc49d245ec5149880cce7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6aa963fdf3970324fa918538ad1578b82426731387c5e61773f5ffe208cfc713", "generation_seconds_batch": 17.601493925903924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.260893821716309, "genuineness": 4.331617832183838, "identity": 0.7199759483337402, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Siborami la prend.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bb1a74d8d7680586ccaae79776b3abed91a4d02565ccc49d245ec5149880cce7.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "bb7f943ddfaa7d196ae1c3fd5779863555ed0b1947a13319416a614a66a3c169", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1ffdc8e5c2853461d66d710a6cd1af483085355f6dde44ba6e80d5748f2c70d", "generation_seconds_batch": 14.738098341040313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.022613525390625, "genuineness": 4.306513786315918, "identity": 0.6710923314094543, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bb7f943ddfaa7d196ae1c3fd5779863555ed0b1947a13319416a614a66a3c169.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "bb96e026cb6bc90f68f9712cd0beb432cb76532c87d695aa5f0fa60a83bfbddb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d25f6b8ded13b3d3dfd0bb637c91f45e935049f0f9fc1a9f923df0f6d68e919", "generation_seconds_batch": 22.10166665702127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8417694568634033, "target_z": 1.0781249617663033, "ce": 6.1894097328186035, "genuineness": 4.105680465698242, "identity": 0.3181634843349457, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bb96e026cb6bc90f68f9712cd0beb432cb76532c87d695aa5f0fa60a83bfbddb.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "bca37d82a25f56a0fd34c44a30a7121a74bb014877afc37a89bd67916cb18e0d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d364fa0854e1a2de11df2cd06ba9fadf8206e9d5c631f98f407bcd766936fabe", "generation_seconds_batch": 16.803584781941026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5173425674438477, "target_z": 1.6328126307592428, "ce": 6.133810997009277, "genuineness": 3.8295347690582275, "identity": 0.32150769233703613, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bca37d82a25f56a0fd34c44a30a7121a74bb014877afc37a89bd67916cb18e0d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "bcc7e1e5d32233a9d60fceaf7b8cbea2057b2a47ef5f9b6f67d908a6d7540816", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81fab7bc97660c338d273344b006bd8ad8296d8f83f174dcbeb1dfe59389338b", "generation_seconds_batch": 17.37158539495431, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.427689552307129, "genuineness": 4.118232727050781, "identity": 0.3220426142215729, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bcc7e1e5d32233a9d60fceaf7b8cbea2057b2a47ef5f9b6f67d908a6d7540816.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "bd07ddd02f0c1aed1c6dbbc1454f862e32e2bcdd63ab6e2240d6555304546f85", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b1caa86e203a3a013a3b381a99d7059af5d5b6848a822eb3980640c4e6c7424", "generation_seconds_batch": 22.33330688904971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.038498878479004, "genuineness": 3.785602569580078, "identity": 0.5474525094032288, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bd07ddd02f0c1aed1c6dbbc1454f862e32e2bcdd63ab6e2240d6555304546f85.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "bd3817ac2b57d81ac991853f80cc152b9add9c2d566caf84ea1988be81c4db5a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7111043aeed3a6ec7c543b385ddfe2d24a62502fa4f2c612492f86efc9f7f6ab", "generation_seconds_batch": 23.341663270955905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.8182125091552734, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bd3817ac2b57d81ac991853f80cc152b9add9c2d566caf84ea1988be81c4db5a.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "bd3bf29f9411297f21d3a312b376a3f812d3a6538699e1501b5299c684c0e4a5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb895275a8a396eaa06cc73d07ecc162e9f498c7926e017875e4cf6a139e765e", "generation_seconds_batch": 27.53396629903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.348263263702393, "genuineness": 3.9676074981689453, "identity": 0.5336710214614868, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bd3bf29f9411297f21d3a312b376a3f812d3a6538699e1501b5299c684c0e4a5.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "bd7598847da9bb59e6af94dc1667337deced83d0a3864c5c15431e2c360f7f72", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "170bc77a602d9015a23ea6cf0925149d499b3c5edbb2359587c035167c5b3bfb", "generation_seconds_batch": 22.649245795910247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 5.756534099578857, "genuineness": 3.860914945602417, "identity": 0.6063463687896729, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bd7598847da9bb59e6af94dc1667337deced83d0a3864c5c15431e2c360f7f72.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "be15a5753e277e897d69cb76bc56dfd92a36cb53f4fd993041deb1ae7fdfa9fb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "317e2dd88332fe4e6088989ab051d223def653218634ee22b9013bd5d24fa9fb", "generation_seconds_batch": 22.174540085019544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.40316104888916, "target_z": 1.5390624808831517, "ce": 6.109982967376709, "genuineness": 3.6412534713745117, "identity": 0.2767493426799774, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/be15a5753e277e897d69cb76bc56dfd92a36cb53f4fd993041deb1ae7fdfa9fb.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "bed5b60d5c1c1fb00de8f8bc227f7f39031191dbaff0b25ddfac8b903a3cce89", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ea611a1e18b54d741b32691479835a56a047d785872a05292d1b784bdd72cfb9", "generation_seconds_batch": 23.910915214917623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5744330883026123, "target_z": 1.679687509940761, "ce": 5.871703147888184, "genuineness": 3.9425034523010254, "identity": 0.7122699618339539, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bed5b60d5c1c1fb00de8f8bc227f7f39031191dbaff0b25ddfac8b903a3cce89.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "bf3e3c827ac094941d944946c85ea2454608a3d13e7566aa848b7eb2307194bd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a7f34a7b3b31ede92aa66238d9a26a8ca2f0bdf7ee2c558d56cc89e4e68b3a8", "generation_seconds_batch": 23.345343463937752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6124935150146484, "target_z": 1.7109374946472824, "ce": 6.340320587158203, "genuineness": 4.105680465698242, "identity": 0.5139728784561157, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bf3e3c827ac094941d944946c85ea2454608a3d13e7566aa848b7eb2307194bd.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "bf8fe025e870a5b2615d0ca133da82d5b5aa1270d3789366c80cfeb421f58c95", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6c0fba47121d4a61b062bc6ce1bf44295dfc8cf877d5b85e1a84387c209171ab", "generation_seconds_batch": 22.047101737000048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.936007261276245, "target_z": 1.9765624625309772, "ce": 6.252951145172119, "genuineness": 4.2751336097717285, "identity": 0.7788500785827637, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bf8fe025e870a5b2615d0ca133da82d5b5aa1270d3789366c80cfeb421f58c95.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "bf9508d25b847f592047aa1321cd6c093d3c22509b4b9ba5766e2b6eea78b8fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f71d69530053c2fc476fce902dd55fa0bb47a471dce7ac799d5a3a138102a7b", "generation_seconds_batch": 16.653242388973013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.260893821716309, "genuineness": 4.362998008728027, "identity": 0.6601201295852661, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bf9508d25b847f592047aa1321cd6c093d3c22509b4b9ba5766e2b6eea78b8fa.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "bf983bc1318d73b66faa03c6c4de8be45e69e418883bf718fdf8d634dc48a3b4", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46efbc5044dc901f7dcbe9f849644172ab6bb4ce2348b6e96e91e1e785dbe4ed", "generation_seconds_batch": 15.561362080043182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.006728172302246, "genuineness": 4.231201171875, "identity": 0.7521402835845947, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/bf983bc1318d73b66faa03c6c4de8be45e69e418883bf718fdf8d634dc48a3b4.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "c00f090db78f64a7612510bac5d2f67e2f09723b98a7622fff17ebe90feed109", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2a9f65d97f3579d554f08202c3e379dd6176f9f3dae841dc3f013b2ac626431", "generation_seconds_batch": 22.97154399496503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.324435234069824, "genuineness": 4.287685871124268, "identity": 0.3100447654724121, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c00f090db78f64a7612510bac5d2f67e2f09723b98a7622fff17ebe90feed109.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "c0b6c46cb48f3d4982d0718387125c766b198fc1981a9964095551112f2b87fe", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2ae730034edfa68422f1a05b14ac6c406f829c5b22983fe2de25092f3a950f7", "generation_seconds_batch": 22.585956795024686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 5.903473377227783, "genuineness": 4.432034492492676, "identity": 0.6711599230766296, "hypothesis": "j'ai trouvé la petite boîte elle était près de la vieille fenêtre nous pouvons regarder ensemble ce qu'elle contient", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c0b6c46cb48f3d4982d0718387125c766b198fc1981a9964095551112f2b87fe.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "c0de91fea38a00cc12b5973cc03a3a393148344355dda48501f896f60495ba2b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82fd6e8c55aabfe08c9ae2f2663c53083be0836424fb4006d6c77fd9de272871", "generation_seconds_batch": 20.887658805935644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.252951145172119, "genuineness": 4.306513786315918, "identity": 0.7065728902816772, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c0de91fea38a00cc12b5973cc03a3a393148344355dda48501f896f60495ba2b.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "c0f34e4767b330820a486ab72440fd19aa00cddd34d5d9e4f058de7ad2ba4ede", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afe71c0f7d29f850cc2a8408512414e6c068ac9280fa79a3a8b29da48f91d22a", "generation_seconds_batch": 15.574978390941396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.229123115539551, "genuineness": 4.068024158477783, "identity": 0.4124118685722351, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c0f34e4767b330820a486ab72440fd19aa00cddd34d5d9e4f058de7ad2ba4ede.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "c103cce6fe70e0ad9ffff4f0aded5ca67f8b0c4a125ffeaf8276047d441aa6bf", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "077654bbf97569d4092301778a7ad0b402d56a2c689e1483e19503bdacbd66f5", "generation_seconds_batch": 17.287164672045037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8417694568634033, "target_z": 1.0781249617663033, "ce": 6.562715530395508, "genuineness": 4.356721878051758, "identity": 0.5301896929740906, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c103cce6fe70e0ad9ffff4f0aded5ca67f8b0c4a125ffeaf8276047d441aa6bf.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "c127496d8dcd0770de09df2c62b8e67d7819670ab7a392840195b7ff2ffe9b7e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19362deb8b632d0dd5c078de18f4f148ac7c3b7cacf5a6a0e7e4c380b7974654", "generation_seconds_batch": 22.33330688904971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 5.871703147888184, "genuineness": 3.9801597595214844, "identity": 0.7110812067985535, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c127496d8dcd0770de09df2c62b8e67d7819670ab7a392840195b7ff2ffe9b7e.mp3" }, { "id": "heldout-fr-3", "language": "fr", "sentences": [ "Quelqu’un a déplacé la petite chaise bleue.", "Elle était ici... près de la grande bibliothèque.", "Nous la remettrons après le déjeuner." ], "text": "Quelqu’un a déplacé la petite chaise bleue. Elle était ici... près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-3", "key": "c13e1aa8dc3b59ea19367ef493ea899879ad8b8c834902c0dacb6bfbfa70eb95", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c518c0074b2f1dd82f8584905c3768405e79311764a182ad52045ccb9648b78", "generation_seconds_batch": 16.959154475014657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.269949436187744, "target_z": 1.4296874365320635, "ce": 6.1894097328186035, "genuineness": 4.231201171875, "identity": 0.5743536353111267, "hypothesis": "Quelqu'un a déplacé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c13e1aa8dc3b59ea19367ef493ea899879ad8b8c834902c0dacb6bfbfa70eb95.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "c15db84656554f644eb8f5bd8e141375915708048309e2c344fde3303043e51c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4d2ed5c011b074886d44b6a4103bac895799731596a9788c880425fb96b7a3d", "generation_seconds_batch": 16.741789147956297, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 5.998785495758057, "genuineness": 3.9299514293670654, "identity": 0.6799425482749939, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c15db84656554f644eb8f5bd8e141375915708048309e2c344fde3303043e51c.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "c1d2afc6eea85a8f8a7e992475266f702a443d1e294e237fc5d28389811381f5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffa3ffc7247cf4d86c5567ed6be30aef90b79c759c6246dea36a26ae1021e1e2", "generation_seconds_batch": 23.88074375106953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.2767791748046875, "genuineness": 3.9676074981689453, "identity": 0.4115784764289856, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c1d2afc6eea85a8f8a7e992475266f702a443d1e294e237fc5d28389811381f5.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "c1e7f374a7c351d1454a2bba6cc0192d8bfa8061da57aa3b9160948abaa279d5", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9eae9826654151cd6e223f0632f86da7fb234f67e619bc7de38701b8dbe31d6f", "generation_seconds_batch": 14.748581159044988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.38003396987915, "genuineness": 3.986435890197754, "identity": 0.5820964574813843, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c1e7f374a7c351d1454a2bba6cc0192d8bfa8061da57aa3b9160948abaa279d5.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "c2f17e7a4e9589089ec5dced1e1ad66c4c21c5db4769b5813a8b4204d2b508f9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b654d4ce21dd38bf7c047fd45b95c182e1abd8580ac3d45d20c225f0880f85f", "generation_seconds_batch": 17.218335265060887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 6.229123115539551, "genuineness": 4.482243061065674, "identity": 0.4855762720108032, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c2f17e7a4e9589089ec5dced1e1ad66c4c21c5db4769b5813a8b4204d2b508f9.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "c36386d5807f883b14d73852210b1e3cc014bfb9eb0ae7c4caa9e15504c7b7b7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f563a763ee33bf4def4233728c554a5bbb71961518d42881d70a18a10e09fa65", "generation_seconds_batch": 21.317286453093402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.117925643920898, "genuineness": 4.111956596374512, "identity": 0.570825457572937, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c36386d5807f883b14d73852210b1e3cc014bfb9eb0ae7c4caa9e15504c7b7b7.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "c38b4f2a1073157853247812823f7f8359eb5c14a7b24234a673d71d705c64a3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "925c928dd57e5b5b49f44e129c5fb798974f40ff9e67bd0eaa6a2934ca003855", "generation_seconds_batch": 16.86874343000818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.205295085906982, "genuineness": 4.281409740447998, "identity": 0.674629807472229, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c38b4f2a1073157853247812823f7f8359eb5c14a7b24234a673d71d705c64a3.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54dccfefe669d19a61845fe1a456381a1d88bae52f4d297479ec8992cc759ed2", "generation_seconds_batch": 17.21819578798022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.252951145172119, "genuineness": 4.344170093536377, "identity": 0.6798954606056213, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "c3a00ccadb29ab3bcacdf590a0b2716fd84bc6dd02a6e874b9d44d9c4ca4ca6e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b28e9e8e88394f2ad6bc2d1c6e3349e3cc1e09ad68dd27840be6ccd211a42ff7", "generation_seconds_batch": 24.670673619024456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 5.9431867599487305, "genuineness": 4.532451152801514, "identity": 0.677419126033783, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro de la casa.", "wer": 0.17647058823529413, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c3a00ccadb29ab3bcacdf590a0b2716fd84bc6dd02a6e874b9d44d9c4ca4ca6e.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "c4576a81a133847d4bd0c7a9b37b139ceb1ec50a2bd33badbb94f0beb5a1ed7f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e4094cf91085216d0e3cb91b1c3c161c1370f5293a8877b3bc5a01a94e1dcaa", "generation_seconds_batch": 15.561362080043182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.536372661590576, "target_z": 1.6484375252342398, "ce": 6.09409761428833, "genuineness": 4.406930446624756, "identity": 0.8395025134086609, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c4576a81a133847d4bd0c7a9b37b139ceb1ec50a2bd33badbb94f0beb5a1ed7f.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba805965f31f26292de1ff26db49a0e953bc0dedaaaea494ae96793bcd5b755e", "generation_seconds_batch": 17.068734186934307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3451571464538574, "target_z": 2.3125000428217404, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.649506688117981, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "c4ee3f0a0dfac709214fc142f2b1a63dea08534552443a441bc132b76431d483", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ace05b3de527d6ecf93234b05acf2c83e67a66e5bb20c8e7320455d9f9672b7a", "generation_seconds_batch": 23.153994086897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.459460258483887, "genuineness": 4.406930446624756, "identity": 0.5295870304107666, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c4ee3f0a0dfac709214fc142f2b1a63dea08534552443a441bc132b76431d483.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "c5fd9cd65baa332c069f80cd6721bd9611a6b6dcd76117f9a4d816b7d737019b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93d54ee47e986241c068deaa2a773485ddf08d53ae1d39df79a755afb58f6b80", "generation_seconds_batch": 22.796518290997483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3555855751037598, "target_z": 1.5000000489391319, "ce": 6.0305562019348145, "genuineness": 4.369274139404297, "identity": 0.5840488076210022, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c5fd9cd65baa332c069f80cd6721bd9611a6b6dcd76117f9a4d816b7d737019b.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "c617ffd97ac2e1166f094ca77b34a78a33b0878e8412bebb304f8aeabeb7bd75", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e84402058454a7cf37a03d0b67be92a4042a5db5158b4218cc2df6bf0cd8b84c", "generation_seconds_batch": 22.195605981047265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.165581703186035, "genuineness": 4.344170093536377, "identity": 0.6278653144836426, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c617ffd97ac2e1166f094ca77b34a78a33b0878e8412bebb304f8aeabeb7bd75.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "c65972027ef3d1ec081e9513c455cc89a3bc9cca035671ef7e888e884d5fa656", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5acf6ceb10b5643a14797a37de536b8383dc6bc90628262b985d90cb89798b7", "generation_seconds_batch": 22.92865046998486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.7792497873306274, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c65972027ef3d1ec081e9513c455cc89a3bc9cca035671ef7e888e884d5fa656.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "c690921595bc0794f4ddade462e51573ed5703972fa6e01354bbe7a7fcb7d824", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "def54af7a18a64b32079a3207c52e1a3e5de700eeb759f80bd980606f90cd684", "generation_seconds_batch": 22.04694971407298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.1020402908325195, "genuineness": 4.168440818786621, "identity": 0.5702750086784363, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c690921595bc0794f4ddade462e51573ed5703972fa6e01354bbe7a7fcb7d824.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "c6cb7d3af8a20e4e049c0d10e1d44b0e0c3727c5817b0deccbbc09614e1ccda7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "15c6febc0cde6328eea01bf9d865dccebf04a980139bc2318381706a17a75028", "generation_seconds_batch": 23.67886245297268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9274054765701294, "target_z": 1.148437476295108, "ce": 6.3641486167907715, "genuineness": 4.262581825256348, "identity": 0.49952432513237, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c6cb7d3af8a20e4e049c0d10e1d44b0e0c3727c5817b0deccbbc09614e1ccda7.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "c7d2dc95737a544e16bcd5e60f35ad80edcbefc2e32749214e070cb55b17a965", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2840f0cbd862498e54b6843ce412411e5fe8967d9269e9ca4cb6d4664c19fc33", "generation_seconds_batch": 21.933430413948372, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.332377910614014, "genuineness": 4.105680465698242, "identity": 0.3259464204311371, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c7d2dc95737a544e16bcd5e60f35ad80edcbefc2e32749214e070cb55b17a965.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "c8121531e85a63bf53a36d59397afa063c0d700d0de4b55515256ff8c7a1199d", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3189bbf0af7f03b6e50470ee5dfa8e7e3ef780b73e351cf88a083078ffd111d", "generation_seconds_batch": 19.039650380029343, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.181467056274414, "genuineness": 4.720732688903809, "identity": 0.7419483065605164, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c8121531e85a63bf53a36d59397afa063c0d700d0de4b55515256ff8c7a1199d.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "c81761f65f930f76053cd759fdfd0e26cbf36a30878258d1414ac99048b43383", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f875a596f0080a6231a1aeb73e4391b62bd14da4e6050428d1ebf0cd0f2f234", "generation_seconds_batch": 17.18378917989321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.181467056274414, "genuineness": 4.331617832183838, "identity": 0.7184231281280518, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c81761f65f930f76053cd759fdfd0e26cbf36a30878258d1414ac99048b43383.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c974ee4829f92873cf3a50437b494213652c7afbf0c1df39d8029667c5bc51c0", "generation_seconds_batch": 16.99337484803982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.006728172302246, "genuineness": 4.262581825256348, "identity": 0.7082582712173462, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b1bff3aca9197c87b043b2469a3244bfdf34011257988e707300170d34ce4aa", "generation_seconds_batch": 15.267286178888753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.348263263702393, "genuineness": 4.319066047668457, "identity": 0.4582602083683014, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "c8c71404fae7bf0e2f433edefbf3b43dd8eac67dce795ab0bea9a51f40d6e6e6", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6409eed9cd742282556e0dae6f062bd627cbe753a92dbc81f0067fc3e1009d02", "generation_seconds_batch": 16.534983063000254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.443574905395508, "genuineness": 4.444586753845215, "identity": 0.5034775733947754, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c8c71404fae7bf0e2f433edefbf3b43dd8eac67dce795ab0bea9a51f40d6e6e6.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "c92ebef113ba8ef1e40fbd28d75c8eee0480abef175621c855c2aa7e15c8c7b3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7a1c5247f7fe7481f333f5023a2736282044bcefb125e154ff87046a581f379", "generation_seconds_batch": 16.43689714395441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.292664527893066, "genuineness": 4.168440818786621, "identity": 0.7291238307952881, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c92ebef113ba8ef1e40fbd28d75c8eee0480abef175621c855c2aa7e15c8c7b3.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "c9380746e9487aad060310732809af32df73de97a8695a4706e4547a52e43a8a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "625de6acc22ad21c782f9d39f4e6529768ea63803983a6a870ca5d1cf8508835", "generation_seconds_batch": 23.21134830499068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.133810997009277, "genuineness": 4.2249250411987305, "identity": 0.79801344871521, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c9380746e9487aad060310732809af32df73de97a8695a4706e4547a52e43a8a.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "c980c39d0110b49ec8f4603314d5aab54d619a26068db22bcaa31f1361a03c27", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a58f5a11e42a1a09d09a36e6e00d1a7e06e7992fbdbb6ddb1c8e3204231d5a44", "generation_seconds_batch": 16.238869221997447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2119455337524414, "target_z": 2.2031249984706522, "ce": 6.165581703186035, "genuineness": 4.369274139404297, "identity": 0.7381769418716431, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c980c39d0110b49ec8f4603314d5aab54d619a26068db22bcaa31f1361a03c27.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "c9db353355c1e44d913a2af4c1d332ea00703a867f746d85e9227524d888b6b0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eeea044726d636d939daa2d5967db63fd3e5d22facd38a2d9f1213dc7a64547f", "generation_seconds_batch": 22.038640736951493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.221180438995361, "genuineness": 4.425758361816406, "identity": 0.40214961767196655, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/c9db353355c1e44d913a2af4c1d332ea00703a867f746d85e9227524d888b6b0.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "ca20c760b392c0963c42bd23361ba62cbb308a48d4952d2286279c0f5c63152f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7bd59eda96b61ed0940e0b17d8978117427fbba220d9e45f9dadd6bff23db607", "generation_seconds_batch": 21.908270057057962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1747984886169434, "target_z": 1.3515625726440237, "ce": 6.213237762451172, "genuineness": 4.300237655639648, "identity": 0.5086443424224854, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ca20c760b392c0963c42bd23361ba62cbb308a48d4952d2286279c0f5c63152f.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "ca619a5bd535a0788236d80789bcdfd14961f630f82616e7bdc7b355e3de3699", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f7b0097b858fe89a2c3e9e2513c55d08a3622a74b5648da4b647ee22808fb68", "generation_seconds_batch": 22.42488304001745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7085579633712769, "target_z": 0.9687500152934787, "ce": 6.229123115539551, "genuineness": 4.419482707977295, "identity": 0.5087622404098511, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ca619a5bd535a0788236d80789bcdfd14961f630f82616e7bdc7b355e3de3699.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "ca8fcf3ea94922bebd4222dd4be21f55207eac3db2730108caf105098321a425", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9819ed9ef8c61cce8eb49696416dd4eeefef35932ec60888d1019b541e613c2c", "generation_seconds_batch": 16.25351162010338, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7846788167953491, "target_z": 1.0312499847065213, "ce": 6.4912309646606445, "genuineness": 4.457139015197754, "identity": 0.39452701807022095, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ca8fcf3ea94922bebd4222dd4be21f55207eac3db2730108caf105098321a425.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "cb00d27be7f1cd5c1fff36dc3de846f913e2d89f379cc5fd8b669035e55a29a5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80d3c23dbf6db0e3790ed088deb1ed4ac9f7d60cf74aa0c6fbfdc447e5c41aea", "generation_seconds_batch": 26.42855558008887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 5.982900142669678, "genuineness": 4.074300289154053, "identity": 0.7270469665527344, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cb00d27be7f1cd5c1fff36dc3de846f913e2d89f379cc5fd8b669035e55a29a5.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "cb9e3c37802a9ed0d10bd6c1741d373d81d517162568b3f16bfc815871e2d9ad", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ce713639ac4b41b835dcfe36a8343e103cd84e5cdf7c008744ff21365843732d", "generation_seconds_batch": 19.11803970800247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.7197157144546509, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cb9e3c37802a9ed0d10bd6c1741d373d81d517162568b3f16bfc815871e2d9ad.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "cc16b3467d8d36859165a4c51784e9bbf863764655b01ad6f2de860b389c6201", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37f8531b63dc150adc4e78a553210d7187931eb534bf61f0460b23c32466ee49", "generation_seconds_batch": 25.043397782952525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.49831223487854, "target_z": 1.6171875405277185, "ce": 6.300607204437256, "genuineness": 4.432034492492676, "identity": 0.5544832944869995, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cc16b3467d8d36859165a4c51784e9bbf863764655b01ad6f2de860b389c6201.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "cc6c75fe20a02c56092d32c655949714f82e6ee26945c3fc70f55171d33388a0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c8ff210320361b4cad26213996ef584153e823090c96eef34cf4dbf2fd14fee", "generation_seconds_batch": 26.094036417081952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 5.967014789581299, "genuineness": 4.369274139404297, "identity": 0.8032765984535217, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cc6c75fe20a02c56092d32c655949714f82e6ee26945c3fc70f55171d33388a0.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "cce8b9c69e973a7f8e3286351af2bddfa9b207628b2ab14fbdf5f3288cad5a36", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34b86b14b3431bdc3d2db64c8314d1e607d385bdab75ef52a00cd0a7839e7bde", "generation_seconds_batch": 22.047101737000048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 5.5738525390625, "genuineness": 4.507347106933594, "identity": 0.7011759281158447, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cce8b9c69e973a7f8e3286351af2bddfa9b207628b2ab14fbdf5f3288cad5a36.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "cd469afad1213039a9cc9b872287d209654331ba027169f59c703fd12b93d5ab", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57e1dd58c5b227918f60ea7910ae14beb528c1838197d7bea53861e677a6165b", "generation_seconds_batch": 21.575078852009028, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 5.458683967590332, "genuineness": 3.7291181087493896, "identity": 0.585618257522583, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cd469afad1213039a9cc9b872287d209654331ba027169f59c703fd12b93d5ab.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "cd50e6ab1ec619e63cfcecece05492ad7beedbb32ace9748ab18808a3ca566c5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "36da616ef7fbeca414ea3acb5a99e8e575b311311d1d2809f35293cfe331426e", "generation_seconds_batch": 23.24653733393643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.631523847579956, "target_z": 1.7265625848788066, "ce": 6.197352409362793, "genuineness": 4.243753433227539, "identity": 0.34625518321990967, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cd50e6ab1ec619e63cfcecece05492ad7beedbb32ace9748ab18808a3ca566c5.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "cd67c395cdb8fb9c2da70a4236d20359793fd2d639a12ece95b35e377da2bcf8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "08883799d4c886d546b049c39bfa44980e1b59b3872046c2a158061029566112", "generation_seconds_batch": 22.340496483957395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.260893821716309, "genuineness": 4.20609712600708, "identity": 0.5603582859039307, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cd67c395cdb8fb9c2da70a4236d20359793fd2d639a12ece95b35e377da2bcf8.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "cda042f76e68876c58dd0826c17e2035eef60ad7f994c4455bb182bfbbff5eb1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33d6ede491eef5facb5e9f6c9a2ec38c61fac65103d0bf6df7d60d3e813e1d74", "generation_seconds_batch": 16.06706341705285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.133810997009277, "genuineness": 4.218649387359619, "identity": 0.7869904041290283, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cda042f76e68876c58dd0826c17e2035eef60ad7f994c4455bb182bfbbff5eb1.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "cdc1b39efc3ed1522faebbd69bded45ce28c6e4947cb06343ffe80e38b9e027e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b59842bae8d004f3877f6580337d9a41ddb11ba0f963aa3bab65741ef0006f0", "generation_seconds_batch": 19.55221467197407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.1894097328186035, "genuineness": 4.080576419830322, "identity": 0.8548699617385864, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cdc1b39efc3ed1522faebbd69bded45ce28c6e4947cb06343ffe80e38b9e027e.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "cdf71cece71acf4b2d8ba57a5e6bb31c37a8cb0b6fd8d95d8414cb07ac9dd706", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "703028a46b90dcc4da2c2a030468e3e2dfb90aed8153e414791c61d589ebd454", "generation_seconds_batch": 21.53497124405112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 6.078212261199951, "genuineness": 3.9801597595214844, "identity": 0.5925766229629517, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cdf71cece71acf4b2d8ba57a5e6bb31c37a8cb0b6fd8d95d8414cb07ac9dd706.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "ce2745ad1ac94d62d47046e88b4a5c0ff6e4b28dc56ceebf127903bbf654b37f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0652e86dccf7643813be92e3c85ef4d5278d9ad1ae1a4819c1f33b91bcd35bad", "generation_seconds_batch": 23.866066505084746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 5.895530700683594, "genuineness": 3.8295347690582275, "identity": 0.6122385859489441, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ce2745ad1ac94d62d47046e88b4a5c0ff6e4b28dc56ceebf127903bbf654b37f.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "ce4ec225bfc59aaf218f9cef2a34e9053eb1f2eb9440671c2ef016fcc2316111", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ef3f7d37bb8958833719f118fade554df579780019f7cda373cf7b8f85d4f42", "generation_seconds_batch": 15.237069964059629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9083752632141113, "target_z": 1.1328124839418474, "ce": 6.316492557525635, "genuineness": 4.507347106933594, "identity": 0.7004401683807373, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ce4ec225bfc59aaf218f9cef2a34e9053eb1f2eb9440671c2ef016fcc2316111.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "cec499ac1030361bb12cc119befc0107aa0d6e959503bd45d96913d252c5d9ba", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0dddba58d56934c932d7d478118c5178067d20db2ffee6b8e789d992078129b", "generation_seconds_batch": 22.889224095037207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.1020402908325195, "genuineness": 4.494794845581055, "identity": 0.65986168384552, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what she contient.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cec499ac1030361bb12cc119befc0107aa0d6e959503bd45d96913d252c5d9ba.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "cedad2b9d6a865f737b0d0020267db4e7912dd9eb41631002590ec67626ac353", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8751b49ef81bb99b28ad320105a6c9dfe1535473075a3f7e734db4a2211584b", "generation_seconds_batch": 23.142422560951672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7457051277160645, "target_z": 1.8203125389983705, "ce": 5.8875885009765625, "genuineness": 3.9676074981689453, "identity": 0.7860856652259827, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cedad2b9d6a865f737b0d0020267db4e7912dd9eb41631002590ec67626ac353.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "cf035e694a27f6b210ae4edcb6ada1d21d66f0e5c377a7f38604af3a915003dc", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50d2d8645d8826efc446409ac91d589dd8bcc1778ce0c5dcca2d514f3cbb5891", "generation_seconds_batch": 17.366462100995705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7418608665466309, "target_z": 0.9960937763812507, "ce": 6.41180419921875, "genuineness": 4.457139015197754, "identity": 0.5118808746337891, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cf035e694a27f6b210ae4edcb6ada1d21d66f0e5c377a7f38604af3a915003dc.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "cf68749a5dd366df4ef6b6cc8bcc868add6b32950b92ae84f6bfd4f15b792907", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e2bdcfb1c20bc3aee7e7ba93001782c953ccd5e77eb9d9ceeefdb564b97fada", "generation_seconds_batch": 16.463702799985185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.157639026641846, "genuineness": 3.8169827461242676, "identity": 0.4206140637397766, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cf68749a5dd366df4ef6b6cc8bcc868add6b32950b92ae84f6bfd4f15b792907.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffa79563be480c3394b7785a98c9d925de4351e7f8d9beb8eae661f2b271a767", "generation_seconds_batch": 16.837343941093422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.340320587158203, "genuineness": 4.306513786315918, "identity": 0.5973995923995972, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "cfae05da7b9f2315027a628cb6ee40ea8562b741c6535b67468e7aa3a84e44cc", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5f5cb3a95c75f70f2c5cfd7f9bf2d2dca4b123be167fded9dfe7a2630e68521e", "generation_seconds_batch": 23.514332983992063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.09409761428833, "genuineness": 3.8671910762786865, "identity": 0.2777574062347412, "hypothesis": "Doch Gorpsem, die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.10526315789473684, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cfae05da7b9f2315027a628cb6ee40ea8562b741c6535b67468e7aa3a84e44cc.mp3" }, { "id": "heldout-nl-0", "language": "nl", "sentences": [ "Je bericht kwam nadat ik vertrokken was.", "Ik las het twee keer... en legde mijn telefoon weg.", "Zeg maar wanneer we kunnen praten." ], "text": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer... en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-0", "key": "cfc457b3a292fa79061d71cd703cfe91a33c3baa351b8f9d502fd443f51b6487", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0549c28e6c079a2c0193cce1683a6f6bf54b5e9d381b66181c226e88fa48be93", "generation_seconds_batch": 25.967078002053313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.403861999511719, "genuineness": 4.218649387359619, "identity": 0.4472057819366455, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/cfc457b3a292fa79061d71cd703cfe91a33c3baa351b8f9d502fd443f51b6487.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "d04b7d2f2b3cb755ed64c2c4d3238554ff26d093773fd0e62aab5f60e5ff546c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4e8f3aa2d1f40bbbe8a7e763b5ac62b35c49928d90ca44b92dd84a462e011ca", "generation_seconds_batch": 15.481736693996936, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.221180438995361, "genuineness": 4.325342178344727, "identity": 0.6618615984916687, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d04b7d2f2b3cb755ed64c2c4d3238554ff26d093773fd0e62aab5f60e5ff546c.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "d0646ec688dd7d60cb104c9a74bf0e4ae64f38a9f373068ad89ab0195d17b706", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11e86c970c8b6b9dbe1a4c85533add6a78e267428edcb3276793a37cd0c09936", "generation_seconds_batch": 26.64413145906292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2033438682556152, "target_z": 1.3750001101130465, "ce": 6.340320587158203, "genuineness": 4.1998209953308105, "identity": 0.4490891695022583, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d0646ec688dd7d60cb104c9a74bf0e4ae64f38a9f373068ad89ab0195d17b706.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "d08435fa85046fd0da159ed9cbdf982e1a206a40639a76d18b637b27673f273c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "d35af7198d7e406c1894a90a9342e4573599c0357c62a4cd2c301d8ff79d26bf", "generation_seconds_batch": 23.418603440979496, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9283188581466675, "target_z": 0.32812503517500097, "ce": 3.465073585510254, "genuineness": 3.5847692489624023, "identity": 0.25728511810302734, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.120000000000001, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d08435fa85046fd0da159ed9cbdf982e1a206a40639a76d18b637b27673f273c.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "d088e95ab3f986d013e534f9e02198ab57b962e364800d99d6edd7020eeadb4a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e78231c7990bd865d898052bc74f8a614b46cc771af1ac4084bc59d370b7078b", "generation_seconds_batch": 17.178703021025285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.032071590423584, "target_z": 1.23437488529891, "ce": 5.77639102935791, "genuineness": 4.557555198669434, "identity": 0.4704148471355438, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d088e95ab3f986d013e534f9e02198ab57b962e364800d99d6edd7020eeadb4a.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "d0bc01820b08472395b0c769be4b82c139020ff32579e40d968a2a30012603e2", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f9ba5675970f80bf4b48f931686db0e7e5610939e3fbed5229e94ad9d6bed4c", "generation_seconds_batch": 16.63660972798243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.157639026641846, "genuineness": 4.15588903427124, "identity": 0.6180176734924316, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d0bc01820b08472395b0c769be4b82c139020ff32579e40d968a2a30012603e2.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "d0c2330882c7b3ed0105c2ac436b25928519d1609b6c7bd2589dffb69cccbb16", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "032f02b84e39d96ed8e8a825ed0de5b9940d516ea495f024a141eecaefa2537c", "generation_seconds_batch": 22.299937922041863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 5.919358730316162, "genuineness": 4.042920112609863, "identity": 0.7620111703872681, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d0c2330882c7b3ed0105c2ac436b25928519d1609b6c7bd2589dffb69cccbb16.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "d0e640432ccd7ee489a8e39ec47ea549e12903ba7f59056b17330fdedb5cdcce", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e3186dd827857e2432f8697c48c1b426bb0170116f69f6866a7730624acbcb2c", "generation_seconds_batch": 18.01796496193856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.324435234069824, "genuineness": 4.15588903427124, "identity": 0.6713401079177856, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d0e640432ccd7ee489a8e39ec47ea549e12903ba7f59056b17330fdedb5cdcce.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "d0fba8192ffa275a67784056df854f09ae655153275747929a8ed3e9f9bc7af0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d9f7ba92ead0673133d1cdc10d26ec4b7ab743c1a5dae49e4d5aa10a08829c1", "generation_seconds_batch": 22.622591705992818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.213237762451172, "genuineness": 4.281409740447998, "identity": 0.47081178426742554, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d0fba8192ffa275a67784056df854f09ae655153275747929a8ed3e9f9bc7af0.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "d117b7784dd86f86d9c511a984ace722a26e6ca576e7624fcdf93a9689560bb9", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86fb6fdde931d43818afadffec06c59f02da8294294067d965676d869ceea006", "generation_seconds_batch": 15.934377996949479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.340320587158203, "genuineness": 4.20609712600708, "identity": 0.5904725790023804, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d117b7784dd86f86d9c511a984ace722a26e6ca576e7624fcdf93a9689560bb9.mp3" }, { "id": "heldout-en-4", "language": "en", "sentences": [ "I wrote the number on the envelope.", "Read it carefully... and check the last two digits.", "Call me when you have a moment." ], "text": "I wrote the number on the envelope. Read it carefully... and check the last two digits. Call me when you have a moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-4", "key": "d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I wrote the number on the envelope.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c796fffa0da7ac65472d40daf164ede2e6f60abd36d61b91979ffeeb84a0df6a", "generation_seconds_batch": 17.013154405052774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4412214756011963, "target_z": 1.570312465589673, "ce": 6.260893821716309, "genuineness": 4.250029563903809, "identity": 0.306157648563385, "hypothesis": "I wrote the number on the envelope. Read it carefully, and check the last two digits. Call me when you have a moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "heldout-nl-2", "language": "nl", "sentences": [ "De weg gaat linksaf na de brug.", "Loop verder... tot aan het kleine houten hek.", "Ik wacht aan de andere kant." ], "text": "De weg gaat linksaf na de brug. Loop verder... tot aan het kleine houten hek. Ik wacht aan de andere kant.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-2", "key": "d12a4ed8436242a089db6e72a52174435bd18b84a8a84d689cf5c1b6d81f6ce8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e1151c5612e000dacb023cd6f2a7bb26fb749b7e34d998c0f2c2aaa913c2d64b", "generation_seconds_batch": 24.472987315966748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.7602651119232178, "hypothesis": "De weg gaat linksaf naar de brug, loop verder tot aan het kleine, houten hek. Ik wacht aan de anderen kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d12a4ed8436242a089db6e72a52174435bd18b84a8a84d689cf5c1b6d81f6ce8.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "d1f829a23c1c30c39648851770331890170f00a19dacd7637cd61566624ba6ba", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bfde8f31e5d242391fa63e10791e92c817a722274fb3bd3815b57bd39237d62c", "generation_seconds_batch": 25.866032514953986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460251808166504, "target_z": 1.5859375558211972, "ce": 6.260893821716309, "genuineness": 4.168440818786621, "identity": 0.676990807056427, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d1f829a23c1c30c39648851770331890170f00a19dacd7637cd61566624ba6ba.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f7b5d57a6f2d9498c4bffd0695fe5e109787035a9427bf46303f18a3ff9bdf7", "generation_seconds_batch": 17.498806288000196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.268836498260498, "genuineness": 4.15588903427124, "identity": 0.7397723197937012, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "d24c9c687710d83378b1a8bf679afe2f6964085d6aad87a4aced41efb73c5138", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b99d5af12a2038802419b69db0dda3a8c94f01bab6ecb6946d4c14987b8bf37", "generation_seconds_batch": 23.46767769707367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.2767791748046875, "genuineness": 4.18726921081543, "identity": 0.6774779558181763, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d24c9c687710d83378b1a8bf679afe2f6964085d6aad87a4aced41efb73c5138.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "d263d15ab1de3c52fb527058e040021ca93f334840d90232fbb5a3e9b1ebb928", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e423a8e21967d59637c291351bd572c5c237048d6ea5dcf830c587d7c05c78dc", "generation_seconds_batch": 18.758449306013063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4564075469970703, "target_z": 0.7617187932040772, "ce": 6.133810997009277, "genuineness": 4.770940780639648, "identity": 0.6605120301246643, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d263d15ab1de3c52fb527058e040021ca93f334840d90232fbb5a3e9b1ebb928.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "d294cf5c9363df5404130d04bfb97c685cafe9425395247de8554d24c61cfe46", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9033e0a6614967791e71555753f9032f6c99a2e3b712e7f8dc5256c3e535c52b", "generation_seconds_batch": 24.491549690021202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.221180438995361, "genuineness": 4.055472373962402, "identity": 0.6461421847343445, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d294cf5c9363df5404130d04bfb97c685cafe9425395247de8554d24c61cfe46.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "d2a561554a0d8c0163cdbacdf94b38a6dea292ccbea302a9a2b2d8be7e6466fa", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55015439b57e4892a7231a8746d7190ba57cd3895309309eba2ca2e21e96de56", "generation_seconds_batch": 23.199069195077755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0796473026275635, "target_z": 1.2734375129994568, "ce": 6.443574905395508, "genuineness": 4.256305694580078, "identity": 0.4819687604904175, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d2a561554a0d8c0163cdbacdf94b38a6dea292ccbea302a9a2b2d8be7e6466fa.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "d2ef37d0134bcdc45e36bac72166401cf0d190036485ce0a0fdff3a2bfe6f8b5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87a92bcb47d336de3cf2465614aeda1d66691249f2e3905342a4d881b456bf4c", "generation_seconds_batch": 21.367885050014593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.09409761428833, "genuineness": 4.344170093536377, "identity": 0.8264551162719727, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er hij samen in kijken.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d2ef37d0134bcdc45e36bac72166401cf0d190036485ce0a0fdff3a2bfe6f8b5.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "d322078cab0c8fe801dc59430b523304854c4fa89a527e068f46ed14f5d57b12", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "edb7ed2478a52faa1df58fc29d2381dfd8512f531b68b1f526ed912d39ec8c30", "generation_seconds_batch": 14.200745025998913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.141753673553467, "genuineness": 4.432034492492676, "identity": 0.4616708755493164, "hypothesis": "Die Tür ist noch offen, ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d322078cab0c8fe801dc59430b523304854c4fa89a527e068f46ed14f5d57b12.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "d33f3f3785c6fdb3a8651c5ebfbba5a8179d9f53ae1eb7e80b2f31e7c0100c13", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e5eb7e0089d10420381e19774800d024d58fdc23a2f3715d97c9394b34d0a2c", "generation_seconds_batch": 24.290733593981713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.897946834564209, "target_z": 1.945312477824456, "ce": 6.09409761428833, "genuineness": 3.961331605911255, "identity": 0.5676878690719604, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d33f3f3785c6fdb3a8651c5ebfbba5a8179d9f53ae1eb7e80b2f31e7c0100c13.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "d345ba25dd57b5f188718cd4f3b91068675c50ea524562fc4b541ce22e62c3b9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f5ef1ff2627b67810705e055d3b2323d3c3534efde20053d80357d306dd86c9", "generation_seconds_batch": 16.821244201040827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 6.356205940246582, "genuineness": 4.080576419830322, "identity": 0.18024396896362305, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d345ba25dd57b5f188718cd4f3b91068675c50ea524562fc4b541ce22e62c3b9.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "d349fd65ac1255308aa2a211a7dfc254b51246c77085b636ae8c466f907dee63", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e682b662269c2904dce3bcc6f05925031a5255fdd748f0332287fd56f90689e4", "generation_seconds_batch": 23.663462784024887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.419747352600098, "genuineness": 4.394378662109375, "identity": 0.7647119760513306, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d349fd65ac1255308aa2a211a7dfc254b51246c77085b636ae8c466f907dee63.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "d37f4ec131c2afc27b4c5a25cd4feaa079368cac7b3a254645cbeb80b961ebb8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61d6193db4b77b04e5f1efdc82e65c5cff9a17b57785344c0fc24c9e6d6c7938", "generation_seconds_batch": 22.951147488085553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.23706579208374, "genuineness": 4.362998008728027, "identity": 0.737841010093689, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d37f4ec131c2afc27b4c5a25cd4feaa079368cac7b3a254645cbeb80b961ebb8.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "d38d32e68693efa68ebef5005a0a8f65883c64485fe1d2e69c8893d8e2f322dc", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b3402898f9d448a25ad1f45c27f43d42b8f00a74279d0f4f76de61089e69c2a", "generation_seconds_batch": 17.018265229999088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.507116317749023, "genuineness": 4.15588903427124, "identity": 0.3625412583351135, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d38d32e68693efa68ebef5005a0a8f65883c64485fe1d2e69c8893d8e2f322dc.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "d3a3785b3df35a7633f982c32f3bb0f67ac813aad5f5c7547beaa7240789e3b2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8f262e81f556e67724003645a2294bfc346e8d4cb59e7bf0483c905396e29630", "generation_seconds_batch": 17.13688526302576, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 5.673136234283447, "genuineness": 3.923675537109375, "identity": 0.7275622487068176, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d3a3785b3df35a7633f982c32f3bb0f67ac813aad5f5c7547beaa7240789e3b2.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "d3bf83366d5415cc0ace88f9f8ac9eedd0ed1bfb5f12bac6e01f54d91829292f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a43c18488fc03a719858bf97446dc25a1102ea28b2c39b8eda944771755a681e", "generation_seconds_batch": 17.901635734946467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.002613067626953, "target_z": 2.0312499847065215, "ce": 6.09409761428833, "genuineness": 4.143336772918701, "identity": 0.6988073587417603, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d3bf83366d5415cc0ace88f9f8ac9eedd0ed1bfb5f12bac6e01f54d91829292f.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "d3ceab250d330b2470e25faa84c01034399f66e5157cd0d6750e92c7b70b3c52", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "126165e1cc2a01f12af314a69ced3c5d17b27367c49332de8048909508af75ad", "generation_seconds_batch": 24.320108307991177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.6873244643211365, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d3ceab250d330b2470e25faa84c01034399f66e5157cd0d6750e92c7b70b3c52.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d52b6c72585b3866d9da26f370ea71184cbfdc08b8d400c0408c35441cd4a06", "generation_seconds_batch": 16.625318131991662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.149696350097656, "genuineness": 4.425758361816406, "identity": 0.7523527145385742, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. Lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "heldout-de-0", "language": "de", "sentences": [ "Deine Nachricht kam, als ich schon weg war.", "Ich las sie zweimal... dann legte ich das Handy weg.", "Sag mir, wann wir reden können." ], "text": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal... dann legte ich das Handy weg. Sag mir, wann wir reden können.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-0", "key": "d4318584df45149130a958e732b5e05dc242feaaae194a565d41a3f191c93a54", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a79dcc77f3242e9faa3c1a0e21f0c310e73b356d7cd56204b7f7ab4c7d254de", "generation_seconds_batch": 16.72048512008041, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3945592641830444, "target_z": 0.7109374946472825, "ce": 6.4912309646606445, "genuineness": 4.469690799713135, "identity": 0.5947595238685608, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d4318584df45149130a958e732b5e05dc242feaaae194a565d41a3f191c93a54.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "d43602820aaf3aa8788814b1bcd143ca2283b91035e812e5ec39c7eee0e3ac32", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3533e79722700480c8ef5c368eb652ba6af036443f055a954567639e64b2cca7", "generation_seconds_batch": 15.558405668009073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.136737823486328, "target_z": 1.3203123921809754, "ce": 6.062326908111572, "genuineness": 4.268857479095459, "identity": 0.2794780731201172, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d43602820aaf3aa8788814b1bcd143ca2283b91035e812e5ec39c7eee0e3ac32.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "d506027207aa6c2ec133eb38d9b9bacf159be18a3d4fb9eb72c20d9d96df13a8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e977ba8ae9dee7c2eff0203fe6f2d8e8da7c3e7bee3abfa904d160637da63a2b", "generation_seconds_batch": 19.19828974502161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.34549808502197266, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d506027207aa6c2ec133eb38d9b9bacf159be18a3d4fb9eb72c20d9d96df13a8.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "d67d2d4994e636aebe43c39b2761fd8c23b859301aa1cd6de287c565498a6f5d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3ac48dfce6b3d303b50b7f2551374d5aee0e02812242b5323faff77a7f23faf", "generation_seconds_batch": 22.125607194029726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.221180438995361, "genuineness": 4.444586753845215, "identity": 0.500084400177002, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d67d2d4994e636aebe43c39b2761fd8c23b859301aa1cd6de287c565498a6f5d.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "d73e4e795d3ba004317432f7d63d4aca6b6f817efb4bc002b804bf981161dc48", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e5f148887a722fae43c4266aaa01ea9adf9f7c3122cb579f2ea1561a393d11a", "generation_seconds_batch": 16.438579705078155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 5.824047088623047, "genuineness": 4.243753433227539, "identity": 0.5273926258087158, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d73e4e795d3ba004317432f7d63d4aca6b6f817efb4bc002b804bf981161dc48.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "d7435053eefc04d52a1b62d7119d03b969548dbfa3f3d63f8668eea8db412d16", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ba80e69a287d3ff28ca2a6a7866eab2f8ec5bdb6a620779a9c1bd9788fae3f8", "generation_seconds_batch": 16.95474880503025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9835829734802246, "target_z": 2.015625090231524, "ce": 6.125868320465088, "genuineness": 4.20609712600708, "identity": 0.5391311645507812, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d7435053eefc04d52a1b62d7119d03b969548dbfa3f3d63f8668eea8db412d16.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "d7b9052aa8758d202388f8e93279593259edc694744a10f352714f99db6e118f", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffdc104c93431fbb4921f0b2210d391b9c6c70f2b23e5cd8578f3ee3f8092807", "generation_seconds_batch": 17.063161210040562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0796473026275635, "target_z": 1.2734375129994568, "ce": 6.300607204437256, "genuineness": 4.193545341491699, "identity": 0.04908186197280884, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d7b9052aa8758d202388f8e93279593259edc694744a10f352714f99db6e118f.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "d7d0ddd53d924ce081024ed5b93b3579484b2b820ce8b6030487c891cfc8315a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62ccabfb24202e61cda3caac796cbe8d2a452c332fb4265417aef677eae42f16", "generation_seconds_batch": 22.92865046998486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.135824680328369, "target_z": 2.1406250290576097, "ce": 6.229123115539551, "genuineness": 4.086852550506592, "identity": 0.6643762588500977, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d7d0ddd53d924ce081024ed5b93b3579484b2b820ce8b6030487c891cfc8315a.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "d7d5ebe4bab5c40ed4f25d7677bce6f32d71099e9e22ab6f7042558f9b8e6459", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "376f6f2bd65251a1257c286558e03062c38d75fd148462c62d533ea71a288f44", "generation_seconds_batch": 21.9562211119337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8607996702194214, "target_z": 1.0937499541195639, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.38560324907302856, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d7d5ebe4bab5c40ed4f25d7677bce6f32d71099e9e22ab6f7042558f9b8e6459.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "d867c780fd5781bf4b71a579e615d8b842e5d163b3d55a2943a7b8851a298e55", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a21120d0f9c0a116dd1487d32ba8ef883d56a7610f64b2d9586c382084b9e0a", "generation_seconds_batch": 24.491549690021202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.157639026641846, "genuineness": 4.306513786315918, "identity": 0.676720142364502, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d867c780fd5781bf4b71a579e615d8b842e5d163b3d55a2943a7b8851a298e55.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "d8a4a29ec30bc9870860afbb61b50e9920de7c6f08f795beea1ead22726b23a5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2c104e2f9c9604c7681b22e44f056f148fc370dfd2f77243264689d6326aa34", "generation_seconds_batch": 21.936667003901675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7561335563659668, "target_z": 1.007812545115762, "ce": 6.308549880981445, "genuineness": 4.457139015197754, "identity": 0.4048888683319092, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d8a4a29ec30bc9870860afbb61b50e9920de7c6f08f795beea1ead22726b23a5.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "d8cb4f51bd8ef36e558393cdd1d4650bc635f0f9ddeaacbb42c27c57aae125ec", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "642b149b983da60ccac6a3cd1b8f0d9c5643b7886bc577aeb06fe2a5fd64ac7e", "generation_seconds_batch": 22.87776218703948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8408560752868652, "target_z": 1.8984374028864104, "ce": 5.593709468841553, "genuineness": 3.6789097785949707, "identity": 0.7099435925483704, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d8cb4f51bd8ef36e558393cdd1d4650bc635f0f9ddeaacbb42c27c57aae125ec.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "d8f921dc9a122dc5ca951f7724c064eaa1774f8a8c9ec1dd35768938dc08efc5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4520e278372d797afa1976fe272a68f21a2578ea286b8e92a729e6a84d3975dc", "generation_seconds_batch": 23.582460697041824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 5.780362129211426, "genuineness": 4.118232727050781, "identity": 0.6307622790336609, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d8f921dc9a122dc5ca951f7724c064eaa1774f8a8c9ec1dd35768938dc08efc5.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "d94cfabc185bda3162b78c627f8e9419e05a2df9cbcb8a2a6300a08caec0db8b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a86a6f922cdc0ab529d74285ae9d288f7dde3e66ce55efa2a7c9148c772d28d9", "generation_seconds_batch": 17.415100169950165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.070132255554199, "target_z": 1.2656250657619583, "ce": 6.157639026641846, "genuineness": 4.545003414154053, "identity": 0.5853546857833862, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt, sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d94cfabc185bda3162b78c627f8e9419e05a2df9cbcb8a2a6300a08caec0db8b.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "d9e983059414b632116ffb142391546d9b76ff1a0d095d32b2b1f269646d46b1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e32a36f78068688d420916e9b45ba801499ee066ab1ee63b449940eca8608250", "generation_seconds_batch": 23.44470277102664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.1020402908325195, "genuineness": 4.344170093536377, "identity": 0.45608875155448914, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/d9e983059414b632116ffb142391546d9b76ff1a0d095d32b2b1f269646d46b1.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "da1e2fc1a8a56c78b22a4cd053cc9215defa4607764cb44c87672636d5d38a35", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f122447000cca7987491d1d3f2ae2d9efc476de0954bc03afaf038b69efc690", "generation_seconds_batch": 16.126218472025357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.403861999511719, "genuineness": 4.419482707977295, "identity": 0.5683451890945435, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/da1e2fc1a8a56c78b22a4cd053cc9215defa4607764cb44c87672636d5d38a35.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "da215d1e6bcec5824978bc4013bfb3939509b017de1251b469c6d742a08320e5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84892dc3525159bc2a1e27485f0f3c3e2381aeb30d263e6a6cf52e9659b8d628", "generation_seconds_batch": 21.367885050014593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3070967197418213, "target_z": 2.281250058115219, "ce": 6.165581703186035, "genuineness": 4.432034492492676, "identity": 0.748489499092102, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/da215d1e6bcec5824978bc4013bfb3939509b017de1251b469c6d742a08320e5.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "da2cc619e7a9d39b3b934e86f4fc9ceda99a765b63889f50dc07117427d3bb99", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1bb7d836099b914ecf728934b487bb6775877963e92f8c67df0f2ad43424f95", "generation_seconds_batch": 27.376497276010923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.467403411865234, "genuineness": 3.722842216491699, "identity": 0.6036047339439392, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, ließ sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/da2cc619e7a9d39b3b934e86f4fc9ceda99a765b63889f50dc07117427d3bb99.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "da399749b806cda645323a255a4c528e04b724c5aa979b5476f012cfb2777131", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "981cf9736c2ffa39ca2daf3c1f287b6d1c5feaac37232f9d856700d85b7af30a", "generation_seconds_batch": 24.1676687739091, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564918041229248, "target_z": 1.6718750627032626, "ce": 4.812349319458008, "genuineness": 4.319066047668457, "identity": 0.5504624843597412, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/da399749b806cda645323a255a4c528e04b724c5aa979b5476f012cfb2777131.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "daf7ef0c7cf39676354336fa5512d0c1b8f57d3d30f43b8e0329c16e108419c4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9749b2af949fdf8274154d33d436889de31081af9ad116d114a5da10ed191ae7", "generation_seconds_batch": 22.17268229695037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193828582763672, "target_z": 1.367187467119021, "ce": 6.348263263702393, "genuineness": 4.218649387359619, "identity": 0.38526421785354614, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/daf7ef0c7cf39676354336fa5512d0c1b8f57d3d30f43b8e0329c16e108419c4.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Your message arrived after I had left.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30c2b9046a5c0e71ad00518a3910762442f8d7b46ecbb863344c01b8b6ffe618", "generation_seconds_batch": 16.994158094050363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4326196908950806, "target_z": 0.7421874793538038, "ce": 6.38003396987915, "genuineness": 4.055472373962402, "identity": 0.3552134037017822, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "db5e01e6d81f1a88ec5fb0568b44cf16d406cb4496501d64b6790620ab085f32", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40e72adc3fef9a7992c2207495ea1da2683010f58683f767640c524068151fea", "generation_seconds_batch": 22.649245795910247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 5.847874641418457, "genuineness": 3.8797430992126465, "identity": 0.6565227508544922, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/db5e01e6d81f1a88ec5fb0568b44cf16d406cb4496501d64b6790620ab085f32.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "db83a46be71dc63ac2921b370706f2ec51a0026d52e48e799440ff91ae89654a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f793425df6c8a4dbf47eef153108a7179833b43c025ae6b1effdf632b6756a36", "generation_seconds_batch": 23.65285486797802, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.1894097328186035, "genuineness": 3.6914620399475098, "identity": 0.351617693901062, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/db83a46be71dc63ac2921b370706f2ec51a0026d52e48e799440ff91ae89654a.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "dbd484a46cdfc3a0d64562c6b335fd959559e2158a20b4802faa1c65fd68b33c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "559e7aa64627f60dfca81fca749583efca94c3bcc9290dc327d1bc7867e1d9f0", "generation_seconds_batch": 24.12653525196947, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 5.927301406860352, "genuineness": 4.507347106933594, "identity": 0.6061508655548096, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dbd484a46cdfc3a0d64562c6b335fd959559e2158a20b4802faa1c65fd68b33c.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "dbdf72588e263b71f78e35c4a37f829f66a30de25ebc70c28fb7d06cdbbbf35c", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "322976a3935d3a4e18eb449e465036b15bef0b69026165b74e2bccd9b1225225", "generation_seconds_batch": 16.59505917003844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.006728172302246, "genuineness": 4.557555198669434, "identity": 0.7377715110778809, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dbdf72588e263b71f78e35c4a37f829f66a30de25ebc70c28fb7d06cdbbbf35c.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "dc30a3ecdbb73ea48548630b1cdf716e7590359e7425065922667658be5613ac", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83c572a73f4649d0166d82866ae69efad06ae28b0af6b5c5bfdc3da019573b3a", "generation_seconds_batch": 16.21571815793868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 6.316492557525635, "genuineness": 4.080576419830322, "identity": 0.439364492893219, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you. When you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dc30a3ecdbb73ea48548630b1cdf716e7590359e7425065922667658be5613ac.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "dc7f17c3a30ecd1dd66f7aa58352c89e9f016c3b49617c2e3d0981d920c7e3f3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75e899741683c117a9f2aeb7d04f296ed420c029cf6ebc3a379595adceeb6396", "generation_seconds_batch": 17.18378917989321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.133810997009277, "genuineness": 4.080576419830322, "identity": 0.7430042028427124, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dc7f17c3a30ecd1dd66f7aa58352c89e9f016c3b49617c2e3d0981d920c7e3f3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "dca843e3a48d7a9cca369526c4255734390bdf4b826661069bdc70f1979e8e12", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5520efd701ace1b6de6a569bc4473c7b9a046ae4bef55ef910001f5fddba55f3", "generation_seconds_batch": 23.23506689397618, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.157639026641846, "genuineness": 4.162164688110352, "identity": 0.7721402645111084, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dca843e3a48d7a9cca369526c4255734390bdf4b826661069bdc70f1979e8e12.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "dcaab0578b2342b4996b93e04e732b53af9a1b244b0ee1955c82d54968b9ac72", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c29f15e50f0c8d4510f3dd6d67cf04ad6d43c81e2bf7832dfa67945f5bbfdc5d", "generation_seconds_batch": 20.88879332097713, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.38003396987915, "genuineness": 3.9676074981689453, "identity": 0.267703115940094, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dcaab0578b2342b4996b93e04e732b53af9a1b244b0ee1955c82d54968b9ac72.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "dcc2a3402a1deae07012b4e7263fb1e883fb9ecc280059000072c43bbe52d260", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bc5bd4b415a2cd0685809f8a34febce0bbc867fa080ecc8a08b0dd4aa5525a6", "generation_seconds_batch": 17.706203128094785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 6.24500846862793, "genuineness": 4.262581825256348, "identity": 0.22798185050487518, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dcc2a3402a1deae07012b4e7263fb1e883fb9ecc280059000072c43bbe52d260.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "dd6d57751e993b406c85215069717900425dda0ec8ee51afc5515d3c5924365b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f578622eecb544440f9c06833a69d445c853bc4a53537a4eb5334b1b971ac956", "generation_seconds_batch": 23.140089018968865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.003526449203491, "target_z": 1.2109375435864143, "ce": 6.38003396987915, "genuineness": 4.375550270080566, "identity": 0.555183470249176, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dd6d57751e993b406c85215069717900425dda0ec8ee51afc5515d3c5924365b.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "dd8b17de2d4c8232cdb29eea8e87ccdb011a75a65a87c936f6e8fb74b5ba9a9e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "54743a656f1faf1977d77da1f0c7c8497e9b41448b65e1086f771747364241ca", "generation_seconds_batch": 16.643371348036453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.173524379730225, "genuineness": 4.344170093536377, "identity": 0.5723882913589478, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dd8b17de2d4c8232cdb29eea8e87ccdb011a75a65a87c936f6e8fb74b5ba9a9e.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "de0c372bd11ea0a88761c74be71fa6bbf31f234d31b3cc2d9f055c610e19b7c4", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "899a261b756ff26c30ba661f77b25ae852224e9b8581e1aeb491eb5323fabb89", "generation_seconds_batch": 16.549172737053595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.24500846862793, "genuineness": 4.024092197418213, "identity": 0.300784707069397, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/de0c372bd11ea0a88761c74be71fa6bbf31f234d31b3cc2d9f055c610e19b7c4.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "de743d5773ca9aaa97e460b64545c3cd4ac07d2ceeba86dbd4e58b911e1957ff", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95e8213d74f8b29693b7b56e11636170e0a36fe2e30a5f374c2d553ea267c2a3", "generation_seconds_batch": 22.52482335397508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 6.1894097328186035, "genuineness": 4.093128204345703, "identity": 0.585310697555542, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/de743d5773ca9aaa97e460b64545c3cd4ac07d2ceeba86dbd4e58b911e1957ff.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "dec2b39f6e53dec4cfa4739941da27833e94d13e8b6cb0e1ff4180f0f61dc04e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e26b4d6cf7c6e5d94389ce5f6b4e5a5bf9037e422519769307876e2eb276641", "generation_seconds_batch": 24.624230314977467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0130414962768555, "target_z": 1.2187499908239128, "ce": 6.395919322967529, "genuineness": 4.306513786315918, "identity": 0.5742380023002625, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dec2b39f6e53dec4cfa4739941da27833e94d13e8b6cb0e1ff4180f0f61dc04e.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "dec523e3f38702ca52aaaabdfb74dd8134aca6e74aa810ce243ae3baa558efec", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ddcdf307f91c95ad4de354117b4976235c79d67bab23e6b54736c1c77cb3833", "generation_seconds_batch": 17.200013309018686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 5.855817794799805, "genuineness": 3.8420867919921875, "identity": 0.6117942333221436, "hypothesis": "He encontrado la cajita. Estaba junto al ventana vieja. Podemos mirar juntos lo que hay.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dec523e3f38702ca52aaaabdfb74dd8134aca6e74aa810ce243ae3baa558efec.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "dee1fa822ebfc3596247d75ab6fb0b59822aba2c285a21cc4bb6f60941ea4ada", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1ab442071a538887814ab142689f2b70682277aa1e671edbc41ed316c7ce992", "generation_seconds_batch": 14.933941408060491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8227392435073853, "target_z": 1.0624999694130426, "ce": 6.38797664642334, "genuineness": 4.356721878051758, "identity": 0.3170894682407379, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dee1fa822ebfc3596247d75ab6fb0b59822aba2c285a21cc4bb6f60941ea4ada.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "dfa1558be6a8d9122e5c783b57df895b6105715af92beec597d7df4a4dc72de3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "238402fc300d808bd6330f89e277b8cd490968c52470aa8afc1217c4537212a2", "generation_seconds_batch": 16.097556525026448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.125868320465088, "genuineness": 4.394378662109375, "identity": 0.5542008876800537, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dfa1558be6a8d9122e5c783b57df895b6105715af92beec597d7df4a4dc72de3.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "dfa3aa08fe2043f80114318c1a36f4d77ff2ed832eaebcdd35a3947140bc2a3a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "018143b44a26bc822c9d65062adc01238ea05017af415e5cc063a5effa3e32a6", "generation_seconds_batch": 23.054461400024593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6895277500152588, "target_z": 0.9531250229402181, "ce": 6.086154937744141, "genuineness": 4.143336772918701, "identity": 0.7107053995132446, "hypothesis": "De deur staat nog open. Ik heb je jas op stoegel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dfa3aa08fe2043f80114318c1a36f4d77ff2ed832eaebcdd35a3947140bc2a3a.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "dffeecc6b1358b3ab726acad423f8bd265cefce8dbd2c88b0c2acc6f471a10b3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12f46dbf894c23b7c9772557d4a2c64c815d41e09987f8d35aecc4c65cfcb3da", "generation_seconds_batch": 21.90743589401245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 5.982900142669678, "genuineness": 4.030367851257324, "identity": 0.7367774844169617, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/dffeecc6b1358b3ab726acad423f8bd265cefce8dbd2c88b0c2acc6f471a10b3.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e053b8e6896d533517d22e0b87a1f03966da80498728a7989c3590a1ae68c2c8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dafff1c077861715d57ac8fe2f93b72450aee1869c228f7b342e5c959327551e", "generation_seconds_batch": 21.040646870969795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.117925643920898, "genuineness": 4.331617832183838, "identity": 0.7106016278266907, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e053b8e6896d533517d22e0b87a1f03966da80498728a7989c3590a1ae68c2c8.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "e0ba86120ce8ae28a8d047caa7199963b87f363e4bf7e56c950e7fce97cd8433", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3320e6b6a455968279831e7c5af5972bc56c9a7ef4c261398716e6b27131d61", "generation_seconds_batch": 16.028332632035017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.5872862339019775, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e0ba86120ce8ae28a8d047caa7199963b87f363e4bf7e56c950e7fce97cd8433.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "e0c7839e6302d0ca7fb91a50b891bb216471fee470d5e6030f4dd1e4269f3228", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f60e266f096016379cc0daa27612996ddfdd07d2beb2d67dfd45bbf612505699", "generation_seconds_batch": 16.971699091023766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.173524379730225, "genuineness": 4.080576419830322, "identity": 0.7169002294540405, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e0c7839e6302d0ca7fb91a50b891bb216471fee470d5e6030f4dd1e4269f3228.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "e0ddfe159cd97316837684c3a3dfe4b039adfb83ad5a0b9c5c5f6bf9e3ecc4ca", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2d47cee603f06590fcc6f70685ebba8de605542829a35b6a165d839c3aab4a7", "generation_seconds_batch": 16.932238746085204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8503713607788086, "target_z": 1.9062500458804361, "ce": 6.1894097328186035, "genuineness": 4.168440818786621, "identity": 0.6446402668952942, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e0ddfe159cd97316837684c3a3dfe4b039adfb83ad5a0b9c5c5f6bf9e3ecc4ca.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e11e6815d291e43401e71cd003058169f6ba7df7df5491dc802068b61da7ffcb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c6ebd6298c5174d83fb64ec86cab9d7297e3885b62f6bebd56e4ef394f3303a", "generation_seconds_batch": 23.090065954020247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193828582763672, "target_z": 1.367187467119021, "ce": 6.268836498260498, "genuineness": 4.494794845581055, "identity": 0.7625046372413635, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e11e6815d291e43401e71cd003058169f6ba7df7df5491dc802068b61da7ffcb.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e156b829d144d2d24558c01ae3d647e74e7cc712cc97f859a3300fd99a017262", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f27a3a45774c8619df243ebfe475988e8b8012e33a47e8d7a58d6d9c91d40efe", "generation_seconds_batch": 15.816915599047206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.252951145172119, "genuineness": 3.810706615447998, "identity": 0.17184044420719147, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e156b829d144d2d24558c01ae3d647e74e7cc712cc97f859a3300fd99a017262.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "e1814ffd8926389a6d23cdc65375aaa703bd8f5cea88374dd8e4d53ce5a023d8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "313aba7a970651225c31a7c749f7303e96648e57c7ce6a6075cb1ccb08bd407f", "generation_seconds_batch": 19.19828974502161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.3314288258552551, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e1814ffd8926389a6d23cdc65375aaa703bd8f5cea88374dd8e4d53ce5a023d8.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "e26232e5840d1fdb2f2bacc35bc752a85ee564d1eb349f2111de75f65553218c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dfea37578f33fd4edde5f863e666dde9630964ee556203ba456792fd14416ca", "generation_seconds_batch": 22.11672960105352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8037090301513672, "target_z": 1.046874977059782, "ce": 6.340320587158203, "genuineness": 4.482243061065674, "identity": 0.4350256621837616, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e26232e5840d1fdb2f2bacc35bc752a85ee564d1eb349f2111de75f65553218c.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "e2824e2d6cab6f4193a952984949424e31977b3c143846812e564df3d70943e3", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78c7baef195a4a9bc0e24423e5efebef316ce83b497b9e8eb1e9c82b21b6bb43", "generation_seconds_batch": 16.64721522293985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.292664527893066, "genuineness": 4.300237655639648, "identity": 0.6827108860015869, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e2824e2d6cab6f4193a952984949424e31977b3c143846812e564df3d70943e3.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "e2a4e2c5976e1df715aa7fc8f300fb80d21f41d1ef681dee5378f26e95f37f9b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a45eba0bcc4b36b8f5e7faba3bcbcade63d9d7e085a420b7b20885db2b1f312d", "generation_seconds_batch": 16.703064661938697, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.24500846862793, "genuineness": 4.344170093536377, "identity": 0.4743085503578186, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e2a4e2c5976e1df715aa7fc8f300fb80d21f41d1ef681dee5378f26e95f37f9b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "e2c80b3644f3ff9b6a87aefafb50e71e66b74365d2d6e869051756e455fe79e2", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "e2c804668524f487b360ef6600e93346779d7b0d45826fc8e07447f9e46c1e46", "generation_seconds_batch": 16.565797150018625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.32612681388855, "target_z": 2.296874952590216, "ce": 6.062326908111572, "genuineness": 3.6914620399475098, "identity": 0.5039204359054565, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e2c80b3644f3ff9b6a87aefafb50e71e66b74365d2d6e869051756e455fe79e2.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "e2daed6e1bee6966ca19b20a0a7a078633aaaa36ad3b34fba7322c9f07a3ebe8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b3826216c52ef8509c830205b000a7302a6bc97252736de902c4685fa2ae375a", "generation_seconds_batch": 24.469419492059387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.157639026641846, "genuineness": 4.162164688110352, "identity": 0.638217568397522, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e2daed6e1bee6966ca19b20a0a7a078633aaaa36ad3b34fba7322c9f07a3ebe8.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e3aad4861e3979749d109e28ef1b33c77c3f897db2482581fc6a9217b6dfcfe9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "937e9f1b6db24df4fa59108b8a54fa890810d6fdbe2eff1a86e3dc6ffe46602b", "generation_seconds_batch": 16.72356389206834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.308549880981445, "genuineness": 4.011539936065674, "identity": 0.3692454695701599, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e3aad4861e3979749d109e28ef1b33c77c3f897db2482581fc6a9217b6dfcfe9.mp3" }, { "id": "heldout-nl-3", "language": "nl", "sentences": [ "Iemand heeft het kleine blauwe stoeltje verplaatst.", "Het stond hier... naast de hoge boekenkast.", "We zetten het na de lunch terug." ], "text": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier... naast de hoge boekenkast. We zetten het na de lunch terug.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-3", "key": "e4cadfc6e94291f969cbfce6f569021205342bacbdd1e8ae32de5755d4f11ecc", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "376665d9feea80c6230dd52bf3e457980e7ebc69e536b68f7614b9b02706a75f", "generation_seconds_batch": 16.238869221997447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.229123115539551, "genuineness": 4.068024158477783, "identity": 0.8043312430381775, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e4cadfc6e94291f969cbfce6f569021205342bacbdd1e8ae32de5755d4f11ecc.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e4cea0365ee5e942550436a6a0e19c89794cef2f21484e2f754e239fba49b994", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d86efc093e2bc54224dbcfdeb753bf6360baa99f689b33015508e0ec827de0c", "generation_seconds_batch": 16.866452172980644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.7494603395462036, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e4cea0365ee5e942550436a6a0e19c89794cef2f21484e2f754e239fba49b994.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e4dc60522716bb2cf791a7d7152eda2858675283eebb6b47ebe0e2e1670ad008", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "424853df67cfc6ba9daf6120f7ddef9add04abb255ed369167d1b16a4a9db7c1", "generation_seconds_batch": 25.30789974797517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.460251808166504, "target_z": 1.5859375558211972, "ce": 6.006728172302246, "genuineness": 4.193545341491699, "identity": 0.7149686813354492, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e4dc60522716bb2cf791a7d7152eda2858675283eebb6b47ebe0e2e1670ad008.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "e514880bb77c77b77923c52e9f098b7b22b17a4fd554544fdb127b88d55b97db", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "712d32e094fb23030592800a08b6a368d316dcfbe4dbd02ee04a73b78ceef0b3", "generation_seconds_batch": 22.622591705992818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.221180438995361, "genuineness": 4.570107460021973, "identity": 0.7046042680740356, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e514880bb77c77b77923c52e9f098b7b22b17a4fd554544fdb127b88d55b97db.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "e5210116596b672b88a9ba44ab18df495e15076a01669c9ccfa84e831b928cd5", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b14d4575013280280067c1c106d16752934b33e05d6487208e1f35d669447260", "generation_seconds_batch": 17.706203128094785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.527770757675171, "target_z": 0.8203124411201071, "ce": 6.332377910614014, "genuineness": 4.444586753845215, "identity": 0.39073556661605835, "hypothesis": "I found the little box. It was beside the old window. We can look inside together,", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e5210116596b672b88a9ba44ab18df495e15076a01669c9ccfa84e831b928cd5.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e53c8d83803b9d8b72a3dbc7067a5f40bdebd6675a98e593622a071ec75cf559", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3678e46bf5510c69d04fbcb7f8401ef96464cfac96ce5673b61e4386a26629d", "generation_seconds_batch": 25.30789974797517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.6870642900466919, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e53c8d83803b9d8b72a3dbc7067a5f40bdebd6675a98e593622a071ec75cf559.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "e54460a4ec6302c8c4f5eac844318346e50fdc391940e16efde16b1acd172069", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "519aeca73242176858e1b5b2aa2a8285eaecaaa8da9ad674524e4c2e5561727e", "generation_seconds_batch": 16.766077021020465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.127223014831543, "target_z": 1.3125001407000039, "ce": 6.006728172302246, "genuineness": 4.281409740447998, "identity": 0.69020015001297, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e54460a4ec6302c8c4f5eac844318346e50fdc391940e16efde16b1acd172069.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "e55301829ed793f5c0e46a49b0ff8531333c7707caaf0fa2b79bdfa60ed7750b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b1db65441b46a1eefbfe3ca4607b8a4127c7a4ade9c0e7c20d971d006f0e19f", "generation_seconds_batch": 16.614415446063504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.365100622177124, "target_z": 1.5078124961766304, "ce": 6.0146708488464355, "genuineness": 4.369274139404297, "identity": 0.5012326240539551, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e55301829ed793f5c0e46a49b0ff8531333c7707caaf0fa2b79bdfa60ed7750b.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e55bed4bb8a871409225d193bd99646cc8c2315510be1b32ff1ee6cd8c4678ef", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5bcf4bc0d9bb8bee7ca9a1f094375f9474d505cf646ecbc95e32e9862d107677", "generation_seconds_batch": 17.002607790986076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.070269584655762, "genuineness": 4.042920112609863, "identity": 0.626760721206665, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e55bed4bb8a871409225d193bd99646cc8c2315510be1b32ff1ee6cd8c4678ef.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "e596cedee2935ffe6efce0e1654d2c50c2a42cc9e1cbc8c76ead58fc4a9fb644", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6838a498f7f09cd6d06002a32e5c53654d2ff0c90238be5779e0e78fbe86c0e", "generation_seconds_batch": 21.499631618964486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269036293029785, "target_z": 2.250000073408698, "ce": 5.804190158843994, "genuineness": 3.835810661315918, "identity": 0.7305024862289429, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e596cedee2935ffe6efce0e1654d2c50c2a42cc9e1cbc8c76ead58fc4a9fb644.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e59899ba7e2789353838d1341493295355ce31b9046461bc3444f976324da054", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d11fb0c5a7fd89446ab5696408faa247389f48ce2b1998e3eb48cb92d3219a4", "generation_seconds_batch": 22.774087387952022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.332377910614014, "genuineness": 4.174716949462891, "identity": 0.38029730319976807, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e59899ba7e2789353838d1341493295355ce31b9046461bc3444f976324da054.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e5a5f6f2b7733767f1b9bbb9c868ae38459b53484bdd28cc0a1b7ae2ab3fbcf4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12c2df0abdf4084df54a59800d4b41264f91570432e963b121fd71152743e22a", "generation_seconds_batch": 25.27030546707101, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7418608665466309, "target_z": 0.9960937763812507, "ce": 6.078212261199951, "genuineness": 3.7793264389038086, "identity": 0.46405452489852905, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e5a5f6f2b7733767f1b9bbb9c868ae38459b53484bdd28cc0a1b7ae2ab3fbcf4.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "e5ab6eb92865421da6137f303f07a22624d71f868d589444978c84e831397fee", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "16c99875366950e121ba8f2df3624aa82c8542c256ac5121c47d67cb8bec5582", "generation_seconds_batch": 25.97485847293865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8884317874908447, "target_z": 1.9375000305869574, "ce": 6.133810997009277, "genuineness": 4.080576419830322, "identity": 0.6201196908950806, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que", "wer": 0.11764705882352941, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e5ab6eb92865421da6137f303f07a22624d71f868d589444978c84e831397fee.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "e5ce44468b7c908f9dec08e7e4ec9303c1ccf3804cce62410586e090c70f8bf2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca4965e9c9bcd061911e69fedcfe17809135a7b481a640068690ea6e588f87dd", "generation_seconds_batch": 22.94669146090746, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.022613525390625, "genuineness": 4.268857479095459, "identity": 0.5985692143440247, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e5ce44468b7c908f9dec08e7e4ec9303c1ccf3804cce62410586e090c70f8bf2.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "e5d172c556b4e4ca3155b4439a30fd7df392bed27bfff521f87b903728f9814f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "832023a4e79b6d03b2e4aca7575889108cdf61a07e1cced57f4a9c3785179cb2", "generation_seconds_batch": 21.90743589401245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 6.117925643920898, "genuineness": 4.256305694580078, "identity": 0.6318484544754028, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e5d172c556b4e4ca3155b4439a30fd7df392bed27bfff521f87b903728f9814f.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "e6207c4598e57a3ee0c6d07278f0d188d23667c1620a7b24c2632f8d340187f3", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38d8637c0b4296f076aa056c9688c4f7174e2c44262af0401885c8bc477352eb", "generation_seconds_batch": 26.63357648998499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0216434001922607, "target_z": 2.0468750749380455, "ce": 6.117925643920898, "genuineness": 3.9801597595214844, "identity": 0.7290969491004944, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e6207c4598e57a3ee0c6d07278f0d188d23667c1620a7b24c2632f8d340187f3.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "e6ee7a3cbc2d44d2e92b84dc98a78bd1c83d2b664b8d121c67f135fe8cbb8778", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5e08c1f07b115bba266ce48f4b02799f17e623f90afeb3c58733f7fc1ba19d4", "generation_seconds_batch": 18.594492787029594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 6.23706579208374, "genuineness": 4.093128204345703, "identity": 0.7737941741943359, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e6ee7a3cbc2d44d2e92b84dc98a78bd1c83d2b664b8d121c67f135fe8cbb8778.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Someone moved the little blue chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "affa9eb24735d1f9b97daced30b05f11fa57c84e959d70f7d7ec10cb4d677fc0", "generation_seconds_batch": 15.267286178888753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.395919322967529, "genuineness": 4.243753433227539, "identity": 0.4124564528465271, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "e789a8aab75cae2805ed3fa3ffb69d68b76980c7b7ed8e3651b10225b932a816", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9649ef4872c4d70b8fd6e3ce91a7f7b317c2e59da69963c99981b0334528d766", "generation_seconds_batch": 21.112544320058078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.869401454925537, "target_z": 1.9218749403554332, "ce": 6.24500846862793, "genuineness": 4.325342178344727, "identity": 0.7169119715690613, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e789a8aab75cae2805ed3fa3ffb69d68b76980c7b7ed8e3651b10225b932a816.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "e7f764ba2b99f9237d591df273fb40508b1ef9ef3a40e3573a3878a15baa8823", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fa77cd4845a2f689c3ad0da2da030317fb8865af5be13fe35f155ab0f1de094", "generation_seconds_batch": 23.84415880695451, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.38003396987915, "genuineness": 4.218649387359619, "identity": 0.3948947489261627, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e7f764ba2b99f9237d591df273fb40508b1ef9ef3a40e3573a3878a15baa8823.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "e8780a011d6fc8ebb87046e6a05f9d0955c32ccb11601428f98d0c0d2621ca02", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "073519817a978932f46b36ce64075a0ffd9e4feca72db7e63f13943775b31e1a", "generation_seconds_batch": 17.37158539495431, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.964552640914917, "target_z": 2.0, "ce": 6.260893821716309, "genuineness": 4.105680465698242, "identity": 0.21958883106708527, "hypothesis": "Someone moved the little blue chair. It used to stand here beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e8780a011d6fc8ebb87046e6a05f9d0955c32ccb11601428f98d0c0d2621ca02.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e8cf5cdfa74d75551e885d61798d1d8f9df2b848c4a06e4c82fc4c573f76cc82", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f48efffaca52c7a6c2b0ff7a47f80b7525e3f2190bf6eaff13e537a8aa6c5a53", "generation_seconds_batch": 16.105830463115126, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.221180438995361, "genuineness": 3.9048471450805664, "identity": 0.32092538475990295, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e8cf5cdfa74d75551e885d61798d1d8f9df2b848c4a06e4c82fc4c573f76cc82.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "e8e5fa1075ec4301a550dc02b7d4c81ae754679ee229eb45f3968d029f66beb5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e52c2e14cebc2cc53be3f76724af756ffef62a008c236932fd63a3e867fc7035", "generation_seconds_batch": 23.95283246494364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.09409761428833, "genuineness": 4.262581825256348, "identity": 0.5339007377624512, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e8e5fa1075ec4301a550dc02b7d4c81ae754679ee229eb45f3968d029f66beb5.mp3" }, { "id": "heldout-en-0", "language": "en", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-0", "key": "e96cee3e1f999fbaccf07b3dc5522225501acd72206766d5926fc872c82e77f9", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Your message arrived after I had left.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c14c80606cc4cdb315fdf1e62e0368217d0152e9e29a575b5f7c012a54738031", "generation_seconds_batch": 18.086886983946897, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.680012583732605, "target_z": 0.9453124778244559, "ce": 6.435632705688477, "genuineness": 3.9048471450805664, "identity": 0.26266103982925415, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e96cee3e1f999fbaccf07b3dc5522225501acd72206766d5926fc872c82e77f9.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "e9a1f7c14bf54986691af3efb8b3ac5d654b3c49f1e4db25e2bd63e490742c14", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e21a82212905bb1e741a9debd1c2ae6384417d8b675448e553abd44e89b0037", "generation_seconds_batch": 16.286382565973327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.229123115539551, "genuineness": 4.074300289154053, "identity": 0.2308771312236786, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e9a1f7c14bf54986691af3efb8b3ac5d654b3c49f1e4db25e2bd63e490742c14.mp3" }, { "id": "heldout-de-3", "language": "de", "sentences": [ "Jemand hat den kleinen blauen Stuhl umgestellt.", "Er stand hier... neben dem hohen Bücherregal.", "Wir stellen ihn nach dem Essen zurück." ], "text": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier... neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-3", "key": "e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71eced1878ce6713a0506de777d03042fd38c286eb56b9572d5f58ad5e55e204", "generation_seconds_batch": 16.837343941093422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.936007261276245, "target_z": 1.9765624625309772, "ce": 6.38797664642334, "genuineness": 4.143336772918701, "identity": 0.49891191720962524, "hypothesis": "Jemand hat den kleinen blauen Stuhl umgestellt. Er stand hier, neben dem hohen Bücherregal. Wir stellen ihn nach dem Essen zurück.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "eaa9d7d83e98ae01a46acacb50ebe60406de77434f45358e4cb791437234b5c0", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "41ee69328fb83bc3be6f273004b2667ffa52e86a87407e9d760d05d5d29a2b44", "generation_seconds_batch": 14.638222874025814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.205295085906982, "genuineness": 3.873466968536377, "identity": 0.5092744827270508, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eaa9d7d83e98ae01a46acacb50ebe60406de77434f45358e4cb791437234b5c0.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "eb0fa53b931bee125bb4fcf2d9a2405051764aa917653d912c0c8ccfa013f596", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03150ec6a677f248e4a88c58d963658ed9e362443f820704cef4ae2cfebf91f8", "generation_seconds_batch": 22.466275697923265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.859886407852173, "target_z": 1.9140624931179346, "ce": 6.117925643920898, "genuineness": 4.143336772918701, "identity": 0.5439199805259705, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb0fa53b931bee125bb4fcf2d9a2405051764aa917653d912c0c8ccfa013f596.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "eb3672d774929c047470e9f9261296e620152e01d6930abec733a294629242ae", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbabe47a96a578db71578bbdc2ee4a41f91081851cee0bfc32ebc82b80058a8a", "generation_seconds_batch": 23.47612225904595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.459338426589966, "target_z": 2.4062499969413045, "ce": 5.704906940460205, "genuineness": 3.7040140628814697, "identity": 0.8358486890792847, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb3672d774929c047470e9f9261296e620152e01d6930abec733a294629242ae.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "eb4782eacea2914f607242c08bab7b950464c5fc09b208a28e35f4a5ffb1dab8", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e7bba086b820875fe2bc2b21792f13e18f7c0e8f52b785c15c40077f348f400", "generation_seconds_batch": 16.27183519001119, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.213237762451172, "genuineness": 4.356721878051758, "identity": 0.4510873556137085, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb4782eacea2914f607242c08bab7b950464c5fc09b208a28e35f4a5ffb1dab8.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "eb57b9c8b304155ee5ed2e5981c31c07bb83faf19577eab195936c8fc2f0d1f1", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2275914b08f379f2545270adb29c1e5391847891d3895ec6cfad9d09052a8fbd", "generation_seconds_batch": 16.852517124963924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.488797187805176, "target_z": 1.60937509329022, "ce": 5.959072113037109, "genuineness": 4.18099308013916, "identity": 0.7299318313598633, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb57b9c8b304155ee5ed2e5981c31c07bb83faf19577eab195936c8fc2f0d1f1.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "eb6d39c145c1e6c20a901c60871b17d2850a7793663466284164839329449946", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "637db924beafe115fdd48c3d856094d5658d3551cc9888792065ce61311e5489", "generation_seconds_batch": 22.951147488085553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3555855751037598, "target_z": 1.5000000489391319, "ce": 6.23706579208374, "genuineness": 4.344170093536377, "identity": 0.8259215950965881, "hypothesis": "The porte is encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Patrion?", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb6d39c145c1e6c20a901c60871b17d2850a7793663466284164839329449946.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "eb9db055f460a088e5044ff71e1da6dbf12bc576b3a9ba878b341c8e15037389", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "526b06c7521775dc236c964369fc577e8e76e1cf7dc45d9bcd4c37ee3ad10fc2", "generation_seconds_batch": 22.124390727025457, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8408560752868652, "target_z": 1.8984374028864104, "ce": 5.911416053771973, "genuineness": 4.369274139404297, "identity": 0.5792847871780396, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eb9db055f460a088e5044ff71e1da6dbf12bc576b3a9ba878b341c8e15037389.mp3" }, { "id": "heldout-de-4", "language": "de", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-4", "key": "ebdab731c74b43419189d5f5b51dee08a531a64ff277d3b97fe736ab3774a697", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aff89c3809c3aa34009af3247a93a6bb85b85116ca28f6fb60441d41e6b0195c", "generation_seconds_batch": 16.345214131986722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.427689552307129, "genuineness": 4.105680465698242, "identity": 0.6687997579574585, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lies sie genau und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ebdab731c74b43419189d5f5b51dee08a531a64ff277d3b97fe736ab3774a697.mp3" }, { "id": "heldout-fr-2", "language": "fr", "sentences": [ "La route tourne à gauche après le pont.", "Continue... jusqu’à la petite barrière en bois.", "Je t’attendrai de l’autre côté." ], "text": "La route tourne à gauche après le pont. Continue... jusqu’à la petite barrière en bois. Je t’attendrai de l’autre côté.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-2", "key": "ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f48db2943858f60076eafa0cc99971c481a4bdcf19f95b5569f3fc36cef2006", "generation_seconds_batch": 15.696578364004381, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1843135356903076, "target_z": 1.3593750198815222, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.8563902378082275, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "heldout-en-1", "language": "en", "sentences": [ "The keys are on the kitchen table.", "I checked the drawer... then looked beside the window.", "Take them with you when you leave." ], "text": "The keys are on the kitchen table. I checked the drawer... then looked beside the window. Take them with you when you leave.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-1", "key": "ec0e943297ee7a23d1fcbb9cb60c9196b0d2bbd0140a91981b99eefbc80a1986", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The keys are on the kitchen table.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fe5d2c8d7671e63b030e3035a7cecfc3b31be61d798dbb814ff01f6949b8c7c", "generation_seconds_batch": 17.018265229999088, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.059703826904297, "target_z": 2.0781250596445666, "ce": 6.348263263702393, "genuineness": 3.9676074981689453, "identity": 0.34937846660614014, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ec0e943297ee7a23d1fcbb9cb60c9196b0d2bbd0140a91981b99eefbc80a1986.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "ec321b180c1096a7f071c7c7a356d012d1ab7af7083ce9f2c22cf9d57653f2a0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0f89341e1ff6d37476947b33c7a4a544142bb1122a7d3fd50dbfacbb9e97d389", "generation_seconds_batch": 17.002607790986076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.046441555023193, "genuineness": 3.8546390533447266, "identity": 0.6002734303474426, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Dis-moi t'entrefamment, dis apprenant.", "wer": 0.2777777777777778, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ec321b180c1096a7f071c7c7a356d012d1ab7af7083ce9f2c22cf9d57653f2a0.mp3" }, { "id": "heldout-nl-1", "language": "nl", "sentences": [ "De sleutels liggen op de keukentafel.", "Ik keek in de la... en daarna naast het raam.", "Neem ze mee als je weggaat." ], "text": "De sleutels liggen op de keukentafel. Ik keek in de la... en daarna naast het raam. Neem ze mee als je weggaat.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-nl-1", "key": "ec40154a7db32007ff2a8582940c98ddbec889cf86ceaef15c846c40e5fccead", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8406a3c315bc0d4969a577c08a049211451577c317ab524a8cb281a2b2a271ee", "generation_seconds_batch": 16.438579705078155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.592550039291382, "target_z": 2.5156250412923926, "ce": 6.586543083190918, "genuineness": 4.21237325668335, "identity": 0.4725971817970276, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de laa en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ec40154a7db32007ff2a8582940c98ddbec889cf86ceaef15c846c40e5fccead.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "ecb9a06ecfc452d5a00fde68c8b4976f44577dbf09d449c9f381ce127f202174", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fb8ca473cffec145caa06cad8d946e7acc91f751565d338244b7fafb3b678a0", "generation_seconds_batch": 16.809752718079835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.1894097328186035, "genuineness": 3.961331605911255, "identity": 0.4213864505290985, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ecb9a06ecfc452d5a00fde68c8b4976f44577dbf09d449c9f381ce127f202174.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "ed5097cfeef5510728dbb6f037736280d770fca7221c8e5f19ce741ecab0837b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6927d89e2c5917341f7003fe861caede0ee671db8587fe9d1464edc0d1b5c49b", "generation_seconds_batch": 15.944371436955407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1747984886169434, "target_z": 1.3515625726440237, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.6856707334518433, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ed5097cfeef5510728dbb6f037736280d770fca7221c8e5f19ce741ecab0837b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "eda0fabd1227b93e9919276627ce849c6a830122f4c2476f84f1c1ef46cdc23c", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd7d53ece0cb613c852b671526a343be366d10dc63c244ac2eec4fdf7f05f816", "generation_seconds_batch": 23.143088513985276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9074618816375732, "target_z": 1.9531249250619545, "ce": 6.205295085906982, "genuineness": 4.193545341491699, "identity": 0.7248156666755676, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eda0fabd1227b93e9919276627ce849c6a830122f4c2476f84f1c1ef46cdc23c.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "eda6dbbd60fd06f38d876e12eac7cc6118523492a0289a178565a2bf90a70cb2", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "831d8fee0d53a2d2466b48cc66baf0485affe0e1ff4c0c73ac945cba548bfc7a", "generation_seconds_batch": 16.216055219992995, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.24500846862793, "genuineness": 4.18726921081543, "identity": 0.6154986619949341, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eda6dbbd60fd06f38d876e12eac7cc6118523492a0289a178565a2bf90a70cb2.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "edcb63b666fc28914dfdc4af32560708f0ddc39c7992d741ff3c99a34a49decb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "230f5a42f83469f4130313d8c0c3350821e45e6a0b169ec452e2091daf7efeb4", "generation_seconds_batch": 15.995794213027693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4897103309631348, "target_z": 0.7890624564135857, "ce": 5.855817794799805, "genuineness": 4.118232727050781, "identity": 0.5200846195220947, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/edcb63b666fc28914dfdc4af32560708f0ddc39c7992d741ff3c99a34a49decb.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "edcc1c6b29e1fd059d7b81537a9394719ce9eb751432c309e8f686f8b54a1b41", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "add7616cb2f19d98d5f4c0663866209c5854019e69cf55e5dbd67b53104cbc2f", "generation_seconds_batch": 25.78215732902754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.793280601501465, "target_z": 1.8593749709423906, "ce": 6.157639026641846, "genuineness": 4.099404335021973, "identity": 0.7291374206542969, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/edcc1c6b29e1fd059d7b81537a9394719ce9eb751432c309e8f686f8b54a1b41.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ee0538e6f56e624b4b3f99c7852c14606f568e8dddac52489e8ae026662f71b3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "413a99723bf2e79d1f14e9ebe93ca2ce94ede0a9827c6c21c57ac95cb4079be7", "generation_seconds_batch": 24.601513013942167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2794647216796875, "target_z": 1.437500079526089, "ce": 6.23706579208374, "genuineness": 3.9550557136535645, "identity": 0.36652159690856934, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ee0538e6f56e624b4b3f99c7852c14606f568e8dddac52489e8ae026662f71b3.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "eeb99422e624cc400053f6124d9f38e5ab865b7ee37cb6767ad436af131937e7", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a1a0c1a4375e0d63ca0a4c9a0456e2574c878d67d5df060d6e69abea3c926e9", "generation_seconds_batch": 23.870291389059275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.173524379730225, "genuineness": 4.168440818786621, "identity": 0.6356023550033569, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eeb99422e624cc400053f6124d9f38e5ab865b7ee37cb6767ad436af131937e7.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "eedb2c741a8c1615c598898899e4d94f94179f7c6f8231928a57cd1a793d9dcd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53b2f7e3779d65f8995deca73c110b3c2c2f70d08c84ac744a914ac8113b4c33", "generation_seconds_batch": 26.475559833925217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.459460258483887, "genuineness": 4.293961524963379, "identity": 0.7196729183197021, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eedb2c741a8c1615c598898899e4d94f94179f7c6f8231928a57cd1a793d9dcd.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "eefe18bbb825b87e5fbd24bcfb1b8ba3cf75a770257c117f438913f204627312", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb17363fc6439ca9de089975ee9efb738f8760b0d69117cb3538419b208bc971", "generation_seconds_batch": 21.690792967914604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9844961166381836, "target_z": 1.19531245335489, "ce": 6.046441555023193, "genuineness": 4.218649387359619, "identity": 0.6091025471687317, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/eefe18bbb825b87e5fbd24bcfb1b8ba3cf75a770257c117f438913f204627312.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "efac52cc910cb6958665455637a1d41dcb904d21de59bc620dcee6b4cbcfde7b", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e122d97086acd38092d526e5fed961e730e9f1e4fe9e1a5b380229ce663941ee", "generation_seconds_batch": 16.88952737499494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.157639026641846, "genuineness": 4.268857479095459, "identity": 0.6834876537322998, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/efac52cc910cb6958665455637a1d41dcb904d21de59bc620dcee6b4cbcfde7b.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "f0e11a2bf6f02f98e033da859ab3e5b13e3e0d2ce9f11fc56a1305e287c8b6d4", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2991d21d920e79845c8bfd688c287a7009c853ee9d69da9efc381e15a44a4411", "generation_seconds_batch": 22.04694971407298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7552201747894287, "target_z": 1.8281249862358693, "ce": 5.550024509429932, "genuineness": 4.124508857727051, "identity": 0.6167958378791809, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen ingeheerd.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f0e11a2bf6f02f98e033da859ab3e5b13e3e0d2ce9f11fc56a1305e287c8b6d4.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "f0ee947c600ea51b871893a78aa2214dd2e87102e8238a71a085fde69f56f0d8", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03ff2b628d66b6a65a27c7559ff71a90e0e21736e067f0dfb6a5f88fad4c114a", "generation_seconds_batch": 23.514332983992063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 5.982900142669678, "genuineness": 3.785602569580078, "identity": 0.5201997756958008, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f0ee947c600ea51b871893a78aa2214dd2e87102e8238a71a085fde69f56f0d8.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "f20c83cc4e46b4bb8b47e6d9068b19a668dfd866cccdd2b521e1fd1ecf789fac", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f95a6536d26cae0251f197442f758efe9a45061242d7fc56c967937042df948", "generation_seconds_batch": 17.08506784995552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.831341028213501, "target_z": 1.8906249556489119, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.6964818239212036, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f20c83cc4e46b4bb8b47e6d9068b19a668dfd866cccdd2b521e1fd1ecf789fac.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f71ed59900e54d4a79d1147ad371d17e80f6d70cd25fa89867a743bd501b3df2", "generation_seconds_batch": 16.9775777120376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.284721851348877, "genuineness": 4.482243061065674, "identity": 0.66130530834198, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "f2ee4d21febf226448d0aa954c79e3f14336fc028a6485243de7b5c11789e2d3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33790b378c56bd55a61b4c8d13b7a72b3bb454636366afafbadf8ba75ba7c58c", "generation_seconds_batch": 17.40934676793404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8417694568634033, "target_z": 1.0781249617663033, "ce": 6.395919322967529, "genuineness": 4.369274139404297, "identity": 0.30342817306518555, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f2ee4d21febf226448d0aa954c79e3f14336fc028a6485243de7b5c11789e2d3.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f30354e4b7d7c8b69e8d414cecd88b6244cf4dce081d3c35be62e78377d0c156", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bb6fdd031efa67a145b03c2b729b9b589086ab02347e7cf2a1fe0ab4cd85110", "generation_seconds_batch": 22.95822083693929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0406734943389893, "target_z": 2.0624999694130426, "ce": 6.165581703186035, "genuineness": 4.055472373962402, "identity": 0.6306725740432739, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f30354e4b7d7c8b69e8d414cecd88b6244cf4dce081d3c35be62e78377d0c156.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "f35a20e8e36d840ad15924b441f988ff4b3be804175f15d97907cac80677bb1e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c61459e353e6bfd2aee29d8be162ebb9cc355a556740696ff0c98c62237da7b1", "generation_seconds_batch": 28.460626468993723, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5554027557373047, "target_z": 1.6640624197092369, "ce": 6.292664527893066, "genuineness": 4.400654315948486, "identity": 0.672078013420105, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f35a20e8e36d840ad15924b441f988ff4b3be804175f15d97907cac80677bb1e.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "f381e7f3a74c48da4f409aaab6404b803097d7d449fcb21301eaff3275726cc1", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84b817d4b8b9521ad0fd199310adbbd9963d4cd050ba28bccd0ed2e2e3d4f1f2", "generation_seconds_batch": 15.558405668009073, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.133810997009277, "genuineness": 4.293961524963379, "identity": 0.43925267457962036, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f381e7f3a74c48da4f409aaab6404b803097d7d449fcb21301eaff3275726cc1.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "f3bd0eb7aa2d78732e55b791187d4ffbf4f157635106edb9efaaeaee5c7bdec9", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b45ae7421e948a20e72bbab1881f452ca694934fe427c0f5c857029dd22347d", "generation_seconds_batch": 27.39961539604701, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.324435234069824, "genuineness": 4.582659721374512, "identity": 0.5540425777435303, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f3bd0eb7aa2d78732e55b791187d4ffbf4f157635106edb9efaaeaee5c7bdec9.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f42ca27eff649f0d3c02c9c54dbf93736dda2097e161cb045c309d0cc0f74473", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5599814e21019cdd601e5ebf87eba6cacbea0055c9c12dec00c2ed398a615050", "generation_seconds_batch": 17.462935196002945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1548547744750977, "target_z": 2.1562499235326067, "ce": 6.23706579208374, "genuineness": 4.068024158477783, "identity": 0.5857639312744141, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f42ca27eff649f0d3c02c9c54dbf93736dda2097e161cb045c309d0cc0f74473.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "f436210d9fee4c77b43d040960354cb2193392f379f60cad3c20e45acaefa48e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d628233ba66aab0b4fae0a7abb842e86c6e0f3f2dc7c940e0a60d869100a096", "generation_seconds_batch": 20.580536823021248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.022556781768799, "target_z": 1.2265626338179385, "ce": 6.24500846862793, "genuineness": 4.406930446624756, "identity": 0.736397385597229, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f436210d9fee4c77b43d040960354cb2193392f379f60cad3c20e45acaefa48e.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "f4f9fcd92a4f0edab28d4bc707f00b36ee4bc14561e7d7cce7013101550d1a46", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3be798ded5a7baa75b68851bb6f913fbc07879a53f26437838dc381efdb0fbc1", "generation_seconds_batch": 17.898199724964797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 6.09409761428833, "genuineness": 4.281409740447998, "identity": 0.6429233551025391, "hypothesis": "I've trouved the petite boat. It was pretty lay fenêtre. We got ensemble what it contained.", "wer": 0.8421052631578947, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f4f9fcd92a4f0edab28d4bc707f00b36ee4bc14561e7d7cce7013101550d1a46.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "f64cfa6325b5f1679a120c68c2e5dc92d7aaf65ba06ec6f97072fa7c086460f0", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5a0137897f25eb0c14888ef25e083b1247674889507c9aea0dc5e2e4b237f563", "generation_seconds_batch": 21.37740915908944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1747984886169434, "target_z": 1.3515625726440237, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.4724310040473938, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f64cfa6325b5f1679a120c68c2e5dc92d7aaf65ba06ec6f97072fa7c086460f0.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f66e203f695680feab38cdd5d56681955ac4cfe0eee14f824586158720d0aa99", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d463d01dc8ac8bf8720cd4ea2e55c446e621d07a6647fe729315c7729eb3280e", "generation_seconds_batch": 15.753662036033347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9844961166381836, "target_z": 1.19531245335489, "ce": 5.808161735534668, "genuineness": 4.582659721374512, "identity": 0.7729694247245789, "hypothesis": "He encontrado la cajita. Estaba junto a Plal. Estaba junto a la ventana vieja. Pu. Podemos mirar juntos lo que hay dentro.", "wer": 0.29411764705882354, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f66e203f695680feab38cdd5d56681955ac4cfe0eee14f824586158720d0aa99.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "f69106306c4c2250f5f8a9de7c58d09e4a39529c51f1ad6b496450617b71c783", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9b03e69a23123963ec3ac48bed6e57e4cefbd5a4871e101e3ddba934aa90efb", "generation_seconds_batch": 24.091849363059737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 5.998785495758057, "genuineness": 4.3504462242126465, "identity": 0.6714580059051514, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f69106306c4c2250f5f8a9de7c58d09e4a39529c51f1ad6b496450617b71c783.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "f6912c803f8945a5f3c9a84eac0e50a94f947c5831a3861d902133ae54cd71cc", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d2c52b06d2c6a52571e9970f9bb15f6a4afa893433a3879a70d0652a66b1924", "generation_seconds_batch": 18.585814098012634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.680012583732605, "target_z": 0.9453124778244559, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.2586773931980133, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f6912c803f8945a5f3c9a84eac0e50a94f947c5831a3861d902133ae54cd71cc.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f6e777056ece44a21e611689938680c549513bd615014a53234af4b09c627d77", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71873c31739a145fd3ebc2811bb2e771910b7c2a8cba397a8ccc6e6a65f86d33", "generation_seconds_batch": 25.907587827998213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.284721851348877, "genuineness": 4.381826400756836, "identity": 0.7229263782501221, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f6e777056ece44a21e611689938680c549513bd615014a53234af4b09c627d77.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "f717cdff5bd67e15c6a00af8f7d4b11713748aff375cb193661c3e31a30b8399", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3d4cddaecd973b918aa63371589d8a81130471c7feda1903209eff7b9bbb3bb", "generation_seconds_batch": 24.195480598020367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.078212261199951, "genuineness": 4.118232727050781, "identity": 0.48944225907325745, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent. Dan.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f717cdff5bd67e15c6a00af8f7d4b11713748aff375cb193661c3e31a30b8399.mp3" }, { "id": "heldout-fr-1", "language": "fr", "sentences": [ "Les clés sont sur la table de la cuisine.", "J’ai regardé dans le tiroir... puis près de la fenêtre.", "Prends-les quand tu partiras." ], "text": "Les clés sont sur la table de la cuisine. J’ai regardé dans le tiroir... puis près de la fenêtre. Prends-les quand tu partiras.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-1", "key": "f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f7b0f1acb6b3f24b46df58de75360bb32ede3fc161aa6f24c59c9d10c0c4f1a", "generation_seconds_batch": 16.755356175010093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1738851070404053, "target_z": 2.1718750137641307, "ce": 5.998785495758057, "genuineness": 4.300237655639648, "identity": 0.7734760046005249, "hypothesis": "Les clés sont sur la table de la cuisine. J'ai regardé dans le tiroir, puis frais de la fenêtre. Prends-les quand tu partiras.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "heldout-es-1", "language": "es", "sentences": [ "Las llaves están en la mesa de la cocina.", "Miré en el cajón... y luego junto a la ventana.", "Llévatelas cuando te vayas." ], "text": "Las llaves están en la mesa de la cocina. Miré en el cajón... y luego junto a la ventana. Llévatelas cuando te vayas.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-1", "key": "f7c518ced95fb822ab1e1a6975f8bc9c6c5d4e12700f6cbc29c31e6c7ed33c9e", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75a31acbfcaf0603b242173d928b3e375691b795711a2bdff21dc81961cda970", "generation_seconds_batch": 22.018885001074523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1167943477630615, "target_z": 2.124999938826085, "ce": 6.1894097328186035, "genuineness": 4.13078498840332, "identity": 0.809995174407959, "hypothesis": "Las llaves están en la mesa de la cocina. Miré en el cajón y luego junto a la ventana. Llévatelas cuando te vayas.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f7c518ced95fb822ab1e1a6975f8bc9c6c5d4e12700f6cbc29c31e6c7ed33c9e.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "f7d3eb99135eb702e0ccc017b023c0aa8f985be396faf69c7a460d4ebd1801c9", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d981663075d97a5e91d7cd1f595d18c4902a0c43386a8bf4ea0bd45bcab2d43", "generation_seconds_batch": 16.52902959403582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.898860216140747, "target_z": 1.125000036704349, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.546252965927124, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f7d3eb99135eb702e0ccc017b023c0aa8f985be396faf69c7a460d4ebd1801c9.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f7e07a4f971c570297a4313c91814c3e8cf4cbfff5bf5c615df79d57577e05f3", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2c768364b61586881c7c9611fb635b9beadcb46897c7b533330d3ede66595e3", "generation_seconds_batch": 24.670673619024456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4212779998779297, "target_z": 2.375000012234783, "ce": 6.0146708488464355, "genuineness": 4.21237325668335, "identity": 0.8278766870498657, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f7e07a4f971c570297a4313c91814c3e8cf4cbfff5bf5c615df79d57577e05f3.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "f849e33067b7f303efc7277cf2816f2fa96721d67d5faa824e1480c13835ad6f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e849dea9fe1cdf71beb15c8379ff71320c5b0c13197b572a88157926afae01e", "generation_seconds_batch": 23.86672034591902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.752092182636261, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f849e33067b7f303efc7277cf2816f2fa96721d67d5faa824e1480c13835ad6f.mp3" }, { "id": "heldout-de-2", "language": "de", "sentences": [ "Die Straße führt hinter der Brücke nach links.", "Geh weiter... bis zu dem kleinen Holztor.", "Ich warte auf der anderen Seite." ], "text": "Die Straße führt hinter der Brücke nach links. Geh weiter... bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-2", "key": "f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74a87f1d256cbcfbf446247cba1c865a73fc9b69947cd7e23357e0bcfff945ca", "generation_seconds_batch": 16.96080718608573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 6.372091293334961, "genuineness": 4.18099308013916, "identity": 0.33843928575515747, "hypothesis": "Die Straße führt hinter der Brücke nach links. Geh weiter bis zum kleinen Holztor. Ich warte auf der anderen Seite.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "f88d97388aa2b88d3c32a2fadb616abddde377e14a5ecabfb1c97544a00edc83", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dba2d2b0072033383ce4f6115d12da4ac6f77e685a7b08d7e4476770daa7a200", "generation_seconds_batch": 23.234571039909497, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.141753673553467, "genuineness": 4.174716949462891, "identity": 0.28409358859062195, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f88d97388aa2b88d3c32a2fadb616abddde377e14a5ecabfb1c97544a00edc83.mp3" }, { "id": "vn_S_TECH_reg-es-1", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-1", "key": "f8a01fc3163b3495c42c6594c6285258a06541716029b0769aaabc52cb8ff9ef", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0295d42598a97d7e579ff4cdfee7e9bd9f5d71facf66eae3f5a70de97047cc63", "generation_seconds_batch": 23.74840944202151, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6220085620880127, "target_z": 1.7187499418847811, "ce": 6.141753673553467, "genuineness": 4.168440818786621, "identity": 0.6424300074577332, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f8a01fc3163b3495c42c6594c6285258a06541716029b0769aaabc52cb8ff9ef.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "f915495c9f256ca855e7af61d170f13936c787dba5dd1827dd7198b96942e003", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1399c033723e6c1857816b104432c213b6eb54a06f4a16dc5f64fb42fd77e5ec", "generation_seconds_batch": 14.411127874045633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.003526449203491, "target_z": 1.2109375435864143, "ce": 6.427689552307129, "genuineness": 4.607763767242432, "identity": 0.6162722110748291, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f915495c9f256ca855e7af61d170f13936c787dba5dd1827dd7198b96942e003.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "f9aee0bbe16804c7e846ea959a332580d46a6f2b34182fd4e1ec885965eb2d09", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d32423584b35fad7a22e6b6e8d69319d7b6d90796a891c528241cb8d74fa5eaa", "generation_seconds_batch": 16.852653230074793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.125868320465088, "genuineness": 3.7416701316833496, "identity": 0.7739796042442322, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/f9aee0bbe16804c7e846ea959a332580d46a6f2b34182fd4e1ec885965eb2d09.mp3" }, { "id": "vn_S_TECH_reg-fr-0", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-0", "key": "fa264aed7d05acd10819b8654e7fb939a60bc048bb4d75742394954778d9260f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7707bb740a240dc29a0fe2648fc78f3c00f14782559aefa5371f86c906a1a3c6", "generation_seconds_batch": 17.215266718994826, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.197352409362793, "genuineness": 3.9425034523010254, "identity": 0.7913661003112793, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fa264aed7d05acd10819b8654e7fb939a60bc048bb4d75742394954778d9260f.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "fac0307d5c33bf043ff32ef650dfa9323ca1effe9852c57c91268ef63f16c86d", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52760106ec45ed0bcdeca9271900dcab8511d72db37e79d05f8e4a618fa3d296", "generation_seconds_batch": 18.652273544925265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6704975366592407, "target_z": 0.9375000305869573, "ce": 6.24500846862793, "genuineness": 3.8420867919921875, "identity": 0.34271541237831116, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fac0307d5c33bf043ff32ef650dfa9323ca1effe9852c57c91268ef63f16c86d.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "facca363b1ba76a9a1b26a792c10aa32fbcc9359aa6756c78ce8eec11101df8f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ebb4668bffe1b8dabfd6563b4695bb61abcb1c123f21c9860fa59d270b0a043", "generation_seconds_batch": 23.495719761936925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4792819023132324, "target_z": 1.6015624502961943, "ce": 6.372091293334961, "genuineness": 4.432034492492676, "identity": 0.7035492658615112, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/facca363b1ba76a9a1b26a792c10aa32fbcc9359aa6756c78ce8eec11101df8f.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "faffed9fde76a1b11b85c0ddc2d32f34ad808d3dd2f2bcd4a406786687098d05", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c8000d79e22c444ef53a80970571c24ab999e276ad263713ff7c6db47136af9", "generation_seconds_batch": 24.50352984399069, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8607996702194214, "target_z": 1.0937499541195639, "ce": 6.300607204437256, "genuineness": 4.3504462242126465, "identity": 0.4042755961418152, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/faffed9fde76a1b11b85c0ddc2d32f34ad808d3dd2f2bcd4a406786687098d05.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "fb1b0ebfde917fc390f5e1f7822378ceecd729dc6904bec8212c2b3dd7e56f78", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9c2d943b978b2c938d766a3f724c030ab77c39f4de45979a14aa5cbbbfe7e04a", "generation_seconds_batch": 24.410422628978267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.735362708568573, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fb1b0ebfde917fc390f5e1f7822378ceecd729dc6904bec8212c2b3dd7e56f78.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "fb75d0895e6bff1be99021883002ae4a52fa21529d402c2ff6827459d3afaee6", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0f108128cc773f093029c1c358bf6ac6745e5d04a38ee26afd7c08a7c14b414", "generation_seconds_batch": 25.169362952001393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269036293029785, "target_z": 2.250000073408698, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.7343429327011108, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fb75d0895e6bff1be99021883002ae4a52fa21529d402c2ff6827459d3afaee6.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "fb9bc27adbaf44bd06347d0823c66f23ab09281e421de5214eed7fe7dac3f1aa", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17379f0c5436652f71b9073ea8ecd51949c1b845f005eddea455b251b03c4444", "generation_seconds_batch": 17.463426877977327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4468923807144165, "target_z": 0.7539062480883152, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.5982095003128052, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fb9bc27adbaf44bd06347d0823c66f23ab09281e421de5214eed7fe7dac3f1aa.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "fba4ad95faf4f6a91efe47b46df0da0232ffd1ec58e38dcc72890d154915e5bb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3e09b5b327a74a92d31b8788043aeeeedc471c1fd93baaf46793b1154a39755", "generation_seconds_batch": 24.410422628978267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.097764253616333, "target_z": 2.109375044351088, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.4423925280570984, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fba4ad95faf4f6a91efe47b46df0da0232ffd1ec58e38dcc72890d154915e5bb.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "fbf80b11348c77a78ea59ac6ed21218017e2afc9098d6ba10b7ab7d5cc92704f", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "131fa03b3a6cda89ae599b9d38fda95c362e4454ecfd78bc60e39f58190aa054", "generation_seconds_batch": 23.550592538085766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.300607204437256, "genuineness": 3.9801597595214844, "identity": 0.3556509017944336, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fbf80b11348c77a78ea59ac6ed21218017e2afc9098d6ba10b7ab7d5cc92704f.mp3" }, { "id": "heldout-fr-4", "language": "fr", "sentences": [ "J’ai écrit le numéro sur l’enveloppe.", "Lis-le bien... et vérifie les deux derniers chiffres.", "Appelle-moi quand tu auras un moment." ], "text": "J’ai écrit le numéro sur l’enveloppe. Lis-le bien... et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-4", "key": "fc50ee5238afe3437f9756b8e1af4e63f73b83246732e0ea3d8f3b1f57e5e134", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e8ddb311021534bb3ea20cd020589c427ebff87b11825502ef330cdd607205e", "generation_seconds_batch": 15.836011463077739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 6.086154937744141, "genuineness": 4.18099308013916, "identity": 0.4111938178539276, "hypothesis": "J'ai écrit le numéro sur l'enveloppe, lis-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fc50ee5238afe3437f9756b8e1af4e63f73b83246732e0ea3d8f3b1f57e5e134.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "fc653e02c9b9d057dae5cb1bcd53349b9b9f57661ba19eaf1b12ee2ab249c696", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cb735eba6f9a937e747e58cc61351076075f3393d0ddb36c8c72501a7189fd2d", "generation_seconds_batch": 23.910915214917623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9169771671295166, "target_z": 1.9609375680559802, "ce": 5.792276382446289, "genuineness": 3.7542223930358887, "identity": 0.6448080539703369, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fc653e02c9b9d057dae5cb1bcd53349b9b9f57661ba19eaf1b12ee2ab249c696.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "fc6cee4af8eca8751add7c7b3f79196d7b611dd082eabd94b46cfdcd5a577568", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1118ef4d62e3c97a3a053967f4da46c8c98f43bfb96afa9427948c30805de05a", "generation_seconds_batch": 23.573315830086358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.181467056274414, "genuineness": 3.7918784618377686, "identity": 0.4352415204048157, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fc6cee4af8eca8751add7c7b3f79196d7b611dd082eabd94b46cfdcd5a577568.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "fcba3b1241b091dd67fcd85cfe83da5440e41559a60e4741894bd1240b5a3734", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78cdb945b544bf6a41e9f944be3898790e37595e26536ffc4bb634a889b82be9", "generation_seconds_batch": 25.163893157965504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.060616970062256, "target_z": 1.2578124227679326, "ce": 6.054384231567383, "genuineness": 4.68307638168335, "identity": 0.573072075843811, "hypothesis": "J'ai trouvé la petite boîte, elle était près de la vieille fenêtre, nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fcba3b1241b091dd67fcd85cfe83da5440e41559a60e4741894bd1240b5a3734.mp3" }, { "id": "vn_S_TECH_reg-de-0", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-0", "key": "fd27b99031b8e1a1dae8ab0c520f1e3f28c5da1ca1bb1d33f83400deefe21c6e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "649b5df420f8d8116b616561eb9f3eb899c8e9c1d27b2f5994f7430fd8eefa09", "generation_seconds_batch": 23.535628710058518, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.165581703186035, "genuineness": 4.519899368286133, "identity": 0.503541886806488, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd27b99031b8e1a1dae8ab0c520f1e3f28c5da1ca1bb1d33f83400deefe21c6e.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "fd3f1393a838d39301f26f21b03e989d2780b0838f55750cf7a686b0db1a6e9b", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7532463f79c2f82b3a0289f63ae8ca0c4abce98adf0fc1ac8fa243d9eb1f317", "generation_seconds_batch": 23.95085292297881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.181467056274414, "genuineness": 3.9927117824554443, "identity": 0.33293938636779785, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd3f1393a838d39301f26f21b03e989d2780b0838f55750cf7a686b0db1a6e9b.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "fd43abfcb006387d40d308a0374bb4cd9c7b023aebbf3c501485c60f685988c5", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71b85268feea14becb1a0788f6e9f2f63ef7f1698cb24f612c368f2f81a7d61e", "generation_seconds_batch": 22.038640736951493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7085579633712769, "target_z": 0.9687500152934787, "ce": 6.316492557525635, "genuineness": 4.2374773025512695, "identity": 0.43138015270233154, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd43abfcb006387d40d308a0374bb4cd9c7b023aebbf3c501485c60f685988c5.mp3" }, { "id": "vn_S_TECH_reg-nl-0", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-0", "key": "fd55f4af83d134a8b3fc6a7140f924022837e67d2d8f9884a78ad4bf13d9e16a", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "eb4612d0565893bc7c87b234bc29c122a32a387a47ed09bfd3641b14f6daedf9", "generation_seconds_batch": 24.6298312949948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.6643034219741821, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd55f4af83d134a8b3fc6a7140f924022837e67d2d8f9884a78ad4bf13d9e16a.mp3" }, { "id": "heldout-es-2", "language": "es", "sentences": [ "La carretera gira a la izquierda tras el puente.", "Sigue... hasta la pequeña puerta de madera.", "Te esperaré al otro lado." ], "text": "La carretera gira a la izquierda tras el puente. Sigue... hasta la pequeña puerta de madera. Te esperaré al otro lado.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-es-2", "key": "fd5c96af72ddc02ddbfe4284c2bc07c6ab33f74790fbe4e9cccff16a5f374d16", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0efd920b1df33327c542ecf0d9b6d845859bc20fd04a46927872254428f4919", "generation_seconds_batch": 19.11803970800247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 6.038498878479004, "genuineness": 4.325342178344727, "identity": 0.8191102147102356, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd5c96af72ddc02ddbfe4284c2bc07c6ab33f74790fbe4e9cccff16a5f374d16.mp3" }, { "id": "vn_S_TECH_reg-de-1", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-de-1", "key": "fd8b1d3fb420e459ee5fecbe6ca22d12eed0499f0edc43895396c0b0179f6d56", "bucket": "vn_S_TECH_reg", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "544582ea55b44a9043141fb31ba1094dd16fb750d7cd395a00c38b49213b697a", "generation_seconds_batch": 18.01796496193856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5078272819519043, "target_z": 1.624999987765217, "ce": 6.3641486167907715, "genuineness": 4.124508857727051, "identity": 0.5615478754043579, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fd8b1d3fb420e459ee5fecbe6ca22d12eed0499f0edc43895396c0b0179f6d56.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "fda7994cd015503a4b99131fef986dcbdd46b04bf69cfcf1150a0c0d7b7f6025", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0fa4ef6b879fb8585414513230b6b821b38d28283ca557603ed35e3ea732405", "generation_seconds_batch": 27.159921295009553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.419747352600098, "genuineness": 4.1998209953308105, "identity": 0.05444635450839996, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fda7994cd015503a4b99131fef986dcbdd46b04bf69cfcf1150a0c0d7b7f6025.mp3" }, { "id": "vn_S_TECH_reg-es-0", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-es-0", "key": "fdc2bdd6a5f289c64fdcfece0cb1467f0bb8f2095019a8498a8f9c597fa07712", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8857e9f401821e796c288dbffe3836b413f4f218c38cf4fbf5e219085ef5a774", "generation_seconds_batch": 26.450942093040794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.593463182449341, "target_z": 1.6953124044157584, "ce": 6.070269584655762, "genuineness": 4.111956596374512, "identity": 0.7139825820922852, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fdc2bdd6a5f289c64fdcfece0cb1467f0bb8f2095019a8498a8f9c597fa07712.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "fde8eda57fa70f1940632cf423e725f8de60a83c9e47d707d6fd0dbb6114dc5e", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db76a5a4afcfc22cf696415e7cdc1275b6e93d8f4f977b208cff439cb421a21e", "generation_seconds_batch": 23.31823790294584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.23706579208374, "genuineness": 4.024092197418213, "identity": 0.4944217801094055, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fde8eda57fa70f1940632cf423e725f8de60a83c9e47d707d6fd0dbb6114dc5e.mp3" }, { "id": "vn_S_TECH_reg-fr-1", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-fr-1", "key": "fdf7b34cee92bdc96acb26faab92c51d72fe523201610f10a54ad9df7280c5fd", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "279efac94d6ebcef8251b64d060c0a4bb67146dcf7ab84fabf85e83b7a7b1ebd", "generation_seconds_batch": 26.81598642806057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 5.808161735534668, "genuineness": 3.8922953605651855, "identity": 0.5305584669113159, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient, si nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.42105263157894735, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/fdf7b34cee92bdc96acb26faab92c51d72fe523201610f10a54ad9df7280c5fd.mp3" }, { "id": "vn_S_TECH_reg-en-0", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-0", "key": "ff11f643f5929aeefba1dc7167527999fc5c48dcd0efe19f7a087a0263f2d581", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2e6038231569da87d6a47fbbd569b3c6fb58b3132c94561d8815b8ec3ebee21", "generation_seconds_batch": 17.397432690951973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.260893821716309, "genuineness": 3.9048471450805664, "identity": 0.4941590428352356, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ff11f643f5929aeefba1dc7167527999fc5c48dcd0efe19f7a087a0263f2d581.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "ff58c156e1f59fc0c11a2bae652aa8a9dfd51d21efae1a97a822afe78cc9417a", "bucket": "vn_S_TECH_reg", "phase": "heldout", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Someone moved the little blue chair.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ada13aabd0239eccccb50447fc17db1dcc387a19bb13fc9c0f6fef7a46c268a0", "generation_seconds_batch": 16.91844222601503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.230975866317749, "target_z": 2.2187500887021763, "ce": 6.300607204437256, "genuineness": 3.823258876800537, "identity": 0.2273573875427246, "hypothesis": "Someone moved the little blue chair. It used to stand here, beside the tall bookshelf. We can put it back after lunch.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ff58c156e1f59fc0c11a2bae652aa8a9dfd51d21efae1a97a822afe78cc9417a.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "ffa80a2b0caf36f2e6bdb8c5ff699262d5a255c06dcfa3122c0c6faa911e8581", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a1114607d91e7d6b44ba3912516658433a8b8dbdac77995e06a3a9126d46dfa", "generation_seconds_batch": 24.10219299700111, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6410388946533203, "target_z": 1.7343750321163052, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.667171061038971, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ffa80a2b0caf36f2e6bdb8c5ff699262d5a255c06dcfa3122c0c6faa911e8581.mp3" }, { "id": "vn_S_TECH_reg-nl-1", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-nl-1", "key": "ffad7ab24c73d6fbd950978acd7218ff405b31dbd9b0467801374190ae61bcfb", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7db036ed20f4a96c7dc41e9010d83b623f315a51dade00437e5ee953a2c6af7d", "generation_seconds_batch": 24.320108307991177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 6.205295085906982, "genuineness": 4.055472373962402, "identity": 0.7529200315475464, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ffad7ab24c73d6fbd950978acd7218ff405b31dbd9b0467801374190ae61bcfb.mp3" }, { "id": "vn_S_TECH_reg-en-1", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "vn_S_TECH_reg-en-1", "key": "ffc858d1b4ff363236ee7736804f3e3d3c0dea7d5e23954f4d80ae005ff53f13", "bucket": "vn_S_TECH_reg", "phase": "grid", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(A patient teacher guiding one learner, with clear conceptual grouping, explanatory emphasis and approachable pedagogical authority. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2254d63793a0271d46cfbdc0bbcdc381c3c3f34193127a5e7b083c15dbd5f061", "generation_seconds_batch": 23.15567300398834, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.3641486167907715, "genuineness": 4.469690799713135, "identity": 0.24623306095600128, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_TECH_reg/ffc858d1b4ff363236ee7736804f3e3d3c0dea7d5e23954f4d80ae005ff53f13.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "76381574f9ba419ed52a79a838029324c44bc376ad4a6d4d166785236376fe9a", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.879830002784729, "target_z": 1.1093750443510881, "ce": 6.2767791748046875, "genuineness": 4.381826400756836, "identity": 0.46470674872398376, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/89871b32690740a670b55dfc.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "4f9011ce0aa3db4e3225840af169f02141e7b29c28b3d1f0f1674e1fedcc01fb", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.813224196434021, "target_z": 1.054687522175544, "ce": 6.1894097328186035, "genuineness": 4.507347106933594, "identity": 0.5290591716766357, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/23017b9fd9d7ba9f42716e91.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "ae626aea0ea8af9e3e5ec0bf25729a5743fb3aec0407808fb2042e82cc05f587", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.2767791748046875, "genuineness": 4.3504462242126465, "identity": 0.44413334131240845, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/c20c8f05cb1ac0049bd5696e.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "7f42512e34d0300113b8b09e520d243e2c130985b888db078cfc869d709c719b", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.165581703186035, "genuineness": 4.281409740447998, "identity": 0.5281144976615906, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/e16adaee0e259258088bf05c.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "da367475157e701c4a7282d84bb710fc938fb57d84e0f5a66a26a67dfd2c784b", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3746156692504883, "target_z": 1.515624943414129, "ce": 6.252951145172119, "genuineness": 4.482243061065674, "identity": 0.5067390203475952, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/b55db4247a45dc5e6e05091a.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "124a2bc3cb9a23904fb240c0d3ed0bc25859d76324872070c644039a02ebd2d7", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193828582763672, "target_z": 1.367187467119021, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.4777320623397827, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/a564f45fa6b8dcb6cfe6abe1.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "734c6a156c754d5aa59822bd2befa14b27ee61a07a0d3ae1ba9821739abe0785", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.327040195465088, "target_z": 1.476562511470109, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.6854772567749023, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/78e3e1ac7d47a85bc8c360f3.mp3" }, { "key": "original-vn_S_TECH_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Die Tür ist noch offen.", "Ich habe deinen Mantel auf den Stuhl gelegt.", "Sag mir, wenn du bereit bist." ], "text": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "7ecc068b97aef1d4425a73db4659a5669b6a6e65b9ecbe9b669cba84c7f4b2c8", "task_id": "vn_S_TECH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Die Tür ist noch offen.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.117925643920898, "genuineness": 4.381826400756836, "identity": 0.5911595821380615, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/889a48b24cf5906f6f19a3cc.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "bcd05ed204c5fb6a7f7ed36fa5f25d4ccd044ec62f558d1b6156a76c83f5b894", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.435632705688477, "genuineness": 4.118232727050781, "identity": 0.3835449814796448, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/8f6942c507f2fff27527e14a.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "5b5e2e798064aaf9a4feb272bf4686bf15744d877eb6bfa166ce3bdf3e8db45c", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.393646001815796, "target_z": 1.5312500336456532, "ce": 6.3641486167907715, "genuineness": 4.024092197418213, "identity": 0.4847492575645447, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/ba522bead73b00bd8bba6ff7.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "848c1cfe9b9f927747df5561208040e8a3f0c5874b7f6ba5380712d70d420209", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.395919322967529, "genuineness": 4.2249250411987305, "identity": 0.4733978509902954, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/a1f9b33e1632b7b985b3e3b9.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "d879bbddd32d8948a41ab582b7e208edafdf68e706f6ba5cd456d19857ec92dc", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.526857376098633, "target_z": 1.6406248822402143, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.6667537689208984, "hypothesis": "Ich habe die kleine Schachtel gefunden, sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/1b5077085fa7b15b9093b86e.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f065d2ba263a4a68a51a3c0cce274d9e0b079c5bf2d17ebe9f3bd2a45e044a99", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5458877086639404, "target_z": 1.6562499724717383, "ce": 6.284721851348877, "genuineness": 4.325342178344727, "identity": 0.29953423142433167, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/f9dd2bc2e3690bd4a5151bda.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b30d268ffb00e0f0875af836751447e5d5cc7edb05a9a6bbde1ffdb6c21c1f0a", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469766855239868, "target_z": 1.5937500030586957, "ce": 6.395919322967529, "genuineness": 4.362998008728027, "identity": 0.5911951065063477, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/f80534b462253a5811fc141c.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "54033fe0101b439e68e95051dbb69f8889d84c0ab514057e6b6ddef2dd900f17", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2509193420410156, "target_z": 1.4140625420570663, "ce": 6.348263263702393, "genuineness": 4.319066047668457, "identity": 0.6165680885314941, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/1aa74cb56fe98b4f1f708a6a.mp3" }, { "key": "original-vn_S_TECH_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "de", "sentences": [ "Ich habe die kleine Schachtel gefunden.", "Sie stand neben dem alten Fenster.", "Wir können gemeinsam hineinschauen." ], "text": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "9b79ced071f41f97df61cddc8c3806835195b6c6a8999e32a54b26f3add8fffb", "task_id": "vn_S_TECH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.324435234069824, "genuineness": 4.218649387359619, "identity": 0.3997112512588501, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/de/7a480c7448c26554f80fba0d.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f7a85d2b1d06b58dbde27f02f9eb70cb0ce69352195eaf986b4ebe54b0b497e7", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.2767791748046875, "genuineness": 4.093128204345703, "identity": 0.4676586091518402, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/f8103a5b911a865f7070c36f.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "80dff3114b5726981b9a0ef24d12bc17b3255d7923b401f9880f6f2690c92879", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.2767791748046875, "genuineness": 3.948779582977295, "identity": 0.17833909392356873, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/857f3f74c46aa7ca100ab00a.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b03dc4b816e5cd425f1f3ebb8cc2934026765d49027de239fd6d7c86fabf660e", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.332377910614014, "genuineness": 4.18099308013916, "identity": 0.21509499847888947, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/26b660ec78c2ad6ed84c9ca3.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "284bbcf74af95042397c93d7b4d72de206819ec400aac60107ef65e91e6df214", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193828582763672, "target_z": 1.367187467119021, "ce": 6.181467056274414, "genuineness": 3.8044304847717285, "identity": 0.2035181224346161, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/95e74f0f393bb6c35d85581f.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "59d1d4f5733ca75bd31c35c85f471209effe437d65050d15647f35b5444cec0e", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.431706428527832, "target_z": 1.5625000183521744, "ce": 6.1894097328186035, "genuineness": 3.8546390533447266, "identity": 0.24115267395973206, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/f09f646b65aa578edd4a1b5f.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b1f853b61c33405a8aec5e708b02770d878745ff9a2bf76eba239c33550f67bd", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1462531089782715, "target_z": 1.328125035175001, "ce": 6.24500846862793, "genuineness": 3.7918784618377686, "identity": 0.38169506192207336, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/46b93f26ea88a31b47e7ef3f.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "432431a68e216467588fe4b891b9f0bc0cb373af3658670107d8b0191dd353e2", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.1894097328186035, "genuineness": 3.7291181087493896, "identity": 0.16862249374389648, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/5980d9799c20f82acbb941fa.mp3" }, { "key": "original-vn_S_TECH_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "The door is still open.", "I put your coat on the chair.", "Tell me when you are ready." ], "text": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "6addd73d0b5a159eaa539a4970e008fc02e0c54bd05abd9fc41d34c1bfd65817", "task_id": "vn_S_TECH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] The door is still open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] I put your coat on the chair.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.221180438995361, "genuineness": 3.923675537109375, "identity": 0.37761205434799194, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/9bcfb243a8c6e7f45bfca153.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "7b74a2bff76876555633045b91a47b79d40d8058d25efa5f62ba21ac3adeb935", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1557681560516357, "target_z": 1.3359374824124994, "ce": 6.38797664642334, "genuineness": 4.250029563903809, "identity": 0.4434809982776642, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/a4b4de1c68ccf31a0556a505.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "0d0c1a23202f04c51480086cdbef4b642a3d5e6f8bd698dd24879770ee10273b", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.746618390083313, "target_z": 1.0, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.31753426790237427, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/e2bf9b85ed01df5776770e16.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "cd53e6c98979b73abd73147c60072e882b50be755b0a62e8cb9a97a03c77b225", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.594376564025879, "target_z": 0.8749999632956512, "ce": 6.467403411865234, "genuineness": 4.444586753845215, "identity": 0.35654985904693604, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/7f85c169537ba8b55523dfa3.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "2ba3b58ebdffc6ca63da18b58da88f45018d83d9c65406d0232be4e7d3c6f56d", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5468010902404785, "target_z": 0.8359375313516313, "ce": 6.308549880981445, "genuineness": 4.432034492492676, "identity": 0.5072733163833618, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/6a5a5d1495f2e425354e6c73.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "390ce5b1a32365763f8d9ed176e44b88207838339fac8afde059a29aef94a9c7", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0511019229888916, "target_z": 1.249999975530434, "ce": 6.260893821716309, "genuineness": 4.331617832183838, "identity": 0.2147260159254074, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/f57e15742bf0b156f888edf2.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b0d2b4f675284be809a4d6540b93e83104264274170729e133b9cb2e2a26a55c", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9559508562088013, "target_z": 1.1718750137641307, "ce": 6.340320587158203, "genuineness": 4.306513786315918, "identity": 0.31329378485679626, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/ec8b629673aff737a526e183.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f3241c530f8979e78611facd608a1ed53ffcc37ecde94ea3c085860f26000e4f", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.403861999511719, "genuineness": 4.218649387359619, "identity": 0.5012325644493103, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/45ac888058bcf5b1db81e97b.mp3" }, { "key": "original-vn_S_TECH_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "en", "sentences": [ "I found the little box.", "It was beside the old window.", "We can look inside together." ], "text": "I found the little box. It was beside the old window. We can look inside together.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "73afa4e3d7d4bb8b189ef02503f13f9f957c02dd3c3d798b8c418b51ec4a9dcb", "task_id": "vn_S_TECH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] I found the little box.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] It was beside the old window.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We can look inside together.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.165283203125, "target_z": 1.3437499296499982, "ce": 6.340320587158203, "genuineness": 4.344170093536377, "identity": 0.33879023790359497, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/en/17f7c505a9b6d025476f6eae.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "cb5de2917e052da30be26e6c635ff099612d37b06252134fcf7c7acb08c4477e", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.926492214202881, "target_z": 1.9687500152934787, "ce": 5.871703147888184, "genuineness": 3.8671910762786865, "identity": 0.6789575815200806, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/b0c05334a4447b99135ca3bb.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "94cfed3011394042eb0dd21a489e1e96168dc3a8186c2307b01c2471d74c56dd", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.44924211502075195, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/7dc4a777d4f2e63d72c07772.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "ff59c31629fe5082c448c2cc159a91a38768bc36aaad635c820b85a480756a50", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192915201187134, "target_z": 2.1874999082391278, "ce": 5.974957466125488, "genuineness": 3.8797430992126465, "identity": 0.679011344909668, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/7d3af3326e8a09bc8f68c6a5.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "4f3c825951bd49d76d8820ca264500fa17094d277f106e00c88692e089d3a378", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9455223083496094, "target_z": 1.9843749097684757, "ce": 6.157639026641846, "genuineness": 3.911123275756836, "identity": 0.5752966403961182, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/550f3a2014156afe2f91085f.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "84576e7d3707359c73e57b614d933388d7a2dd420beeddd210a70c9ee831a5cf", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7742505073547363, "target_z": 1.8437500764673933, "ce": 5.839932441711426, "genuineness": 4.005263805389404, "identity": 0.580756664276123, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/d696715df8f353706a71a827.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f64663cd62589ded5ce01b5cdda1780ffd96d819a355f680c33e9fb671b344ab", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.24500846862793, "genuineness": 4.143336772918701, "identity": 0.6249959468841553, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/e4f5a241002c35483a3659e1.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "16d1f033ee2bb5faf3294501e1112df87bf97a21222bf25adacc46719af5df7f", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.149696350097656, "genuineness": 4.105680465698242, "identity": 0.4383297264575958, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/35bb94814319cddfc31b8e78.mp3" }, { "key": "original-vn_S_TECH_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "La puerta sigue abierta.", "He dejado tu abrigo sobre la silla.", "Dime cuándo estás listo." ], "text": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f9a4afab0c2c6653d90729c3bc3317b48d9a85347b734cc53eceb2bbeb556956", "task_id": "vn_S_TECH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.4789046049118042, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/15886e04cbc3c974eb5dbe16.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "1fdd54571493a07a421567bbd511dad152eb341762bc476d9dd8e8608551c6a2", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5554027557373047, "target_z": 1.6640624197092369, "ce": 6.197352409362793, "genuineness": 4.3881025314331055, "identity": 0.5879944562911987, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/28e8e318d3285b1f5ef5c5e7.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "398ad47bca98a5b4c7e8ff281a4df02c43d920b60f9a18f986589c33b86e6132", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0787339210510254, "target_z": 2.093749954119564, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.7695381045341492, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/4905d9c2f3b3ac6113ce2240.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "8119e66b0815acf32759afc0b0098cf8d53336bfee50e4980231015ac67b96b1", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.1020402908325195, "genuineness": 4.557555198669434, "identity": 0.6513330936431885, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/783f4a7748c05639a8330be0.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f89d25f915068ea7ad7c9ce0ba684a64cbd8ed1ae2f3833a3bf4a33df2d81991", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7361900806427, "target_z": 1.812500091760872, "ce": 6.133810997009277, "genuineness": 4.482243061065674, "identity": 0.6218106150627136, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/0ec80f0913bc21cec23714b5.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "14d938712e2d799fcd7ccf058803e67f54b93c41a1ee07ad1cb5bff6a19dd602", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.336555242538452, "target_z": 1.4843749587076076, "ce": 6.078212261199951, "genuineness": 4.519899368286133, "identity": 0.6373867988586426, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/5c548bdae5eebd8203fafc03.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "5f0787c4a667458847419aec244e2de2ddc488b404a99894f12083907d7808f6", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 5.95112943649292, "genuineness": 4.306513786315918, "identity": 0.6685311794281006, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/6bf68f2e59f2094a8359238c.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "eca55bc5fc8465c8666d6f44c9f34b685caf32b2bdd75b0fdf446e439e1c91a4", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2889797687530518, "target_z": 1.4453125267635878, "ce": 5.967014789581299, "genuineness": 4.570107460021973, "identity": 0.71234130859375, "hypothesis": "He encontrado la. He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/6b6428d8dba56a3370bac526.mp3" }, { "key": "original-vn_S_TECH_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "es", "sentences": [ "He encontrado la cajita.", "Estaba junto a la ventana vieja.", "Podemos mirar juntos lo que hay dentro." ], "text": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "ea1190b42ca441571164e30340dcc9db2cee92b52a831651c46f0ee3dc47ef45", "task_id": "vn_S_TECH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.149696350097656, "genuineness": 4.645420074462891, "identity": 0.520035445690155, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/es/6e7de0de439ba3a20ec6c580.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "0549877406a4d6c0766fb5226ff4482eebaa66fe858e39df54aaa12b7443099b", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6790993213653564, "target_z": 1.7656250168228265, "ce": 6.1894097328186035, "genuineness": 4.356721878051758, "identity": 0.638276994228363, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/e53f0d7b973ae96c892f0b62.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "63f1c1e5cc961975b872eff794c9d104539a58337e4584b68836216dae1688d7", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8218259811401367, "target_z": 1.8828125084114133, "ce": 6.181467056274414, "genuineness": 4.193545341491699, "identity": 0.6597362756729126, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/4257e6a1aa4b30f33fa2b9eb.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "1b6c43d6870d528e19a3ed1bfce0fd478d0773be75d94917c390727a723f5c6d", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.125868320465088, "genuineness": 4.532451152801514, "identity": 0.6555423736572266, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/ef6d25c029b4b07ff0166e84.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "c4d230bab79a9ec6672398de3cebc3ead3460b6a35879543bcc9fee470beec1c", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1081924438476562, "target_z": 1.2968748547119526, "ce": 6.300607204437256, "genuineness": 4.507347106933594, "identity": 0.436669260263443, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Hassem.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/8e63f4b9e7c89cd20681b6da.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "880a78f60c965e4f085abdaa0fc7bbc952ba4da93aa76145d6da97722c3b808a", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0891623497009277, "target_z": 1.2812499602369554, "ce": 6.125868320465088, "genuineness": 4.319066047668457, "identity": 0.741683840751648, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/32ef5624312bab863a594cf2.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "92007f7465483b015a7ab00a1f56f41b8d346bc20ca07ef7b353bcfe47bcb3d2", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5839481353759766, "target_z": 1.6874999571782596, "ce": 6.23706579208374, "genuineness": 4.306513786315918, "identity": 0.7493004202842712, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/63bbe829dd4e8a4fa78a18b5.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "6e2881d77f33ef9443cf473cecfd88246aeddd3b9602b84b01d9ad1b204db25a", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5087406635284424, "target_z": 0.80468754664511, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.5940964221954346, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/363212f171c44f982fd36570.mp3" }, { "key": "original-vn_S_TECH_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "La porte est encore ouverte.", "J’ai posé ton manteau sur la chaise.", "Dis-moi quand tu es prêt." ], "text": "La porte est encore ouverte. J’ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "22ac3f126f3333535e83092fee77084ff106691afec55f0b2816798f172d8421", "task_id": "vn_S_TECH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La porte est encore ouverte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4126763343811035, "target_z": 1.5468751238771772, "ce": 6.1894097328186035, "genuineness": 4.1998209953308105, "identity": 0.8000507950782776, "hypothesis": "The port is encouraged. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt, et on s'est referdurane.", "wer": 0.5, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/0edf103a042e4c5c292160af.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "060575c4cb147dd07bf671d7e3bd0cd3b2fcedb872412a8ede3d756b04c4cc9a", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2223739624023438, "target_z": 1.3906250045880435, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.7219475507736206, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/37e6760e8a16b52e103c74f2.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "7f3050bb67b0085dda7f1f0f5c1b93b8e04047c86de40aaee95af20a65b91b34", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.149696350097656, "genuineness": 4.469690799713135, "identity": 0.7444130182266235, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/626fc0acee3da287be387b75.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "12373fc732ceac0f309a9c33d96d414c28f244e07755b99eb1b3b7b32e934800", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.133810997009277, "genuineness": 4.124508857727051, "identity": 0.5193242430686951, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/74263a41266e093f96502357.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "9c7fd31f77ed7655562967eae094209124e960b1c6f2a6ba3eb16a8ff67607d3", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9749810695648193, "target_z": 1.1875000061173915, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.6140726804733276, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/8264ac8c345f705ec6d44b3e.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "96ec73830ec56f3829f6c4547e2d7b61cecd40b97f098a730e4bbea0d29c0473", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9940112829208374, "target_z": 1.2031249984706522, "ce": 6.09409761428833, "genuineness": 4.319066047668457, "identity": 0.7342069745063782, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/34d207fb169ad0f421cfd988.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "edb7587cf8543c8ac87baef1cd8bb7b12eb19c7ad019b417d78717f247216e4f", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.613406777381897, "target_z": 0.8906249556489119, "ce": 6.062326908111572, "genuineness": 4.545003414154053, "identity": 0.748864471912384, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/20a943f7bf1b603171d0e81e.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "9adf979774edc88073fa24ace644b369a6a1a67567f9d7b212ff277ab5c8b28b", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.694285273551941, "target_z": 0.9570312465589673, "ce": 6.046441555023193, "genuineness": 4.444586753845215, "identity": 0.6987179517745972, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/30b38c93b92eed2c6b794290.mp3" }, { "key": "original-vn_S_TECH_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "fr", "sentences": [ "J’ai trouvé la petite boîte.", "Elle était près de la vieille fenêtre.", "Nous pouvons regarder ensemble ce qu’elle contient." ], "text": "J’ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu’elle contient.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "0bef7cf963e97d573c9ab5379de8257c2a4152f9d7d237251eb9d5e1eafd3746", "task_id": "vn_S_TECH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.5245376229286194, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/fr/d3b46e941dfc459ca761afd5.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b35c23caeb6ad03d9a77daeacfe7d528d8b311844bc947b120a36e811f9f6592", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.26043438911438, "target_z": 1.421874989294565, "ce": 6.197352409362793, "genuineness": 4.457139015197754, "identity": 0.3272533416748047, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/9b4c3fa3be5d9a62e83bd000.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "bf488e98c535fdcc8b7a8bbc9703d66d992744368af311c35375b9ba1942a397", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.698129415512085, "target_z": 1.7812499112978237, "ce": 6.133810997009277, "genuineness": 3.9801597595214844, "identity": 0.4563969373703003, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/fdf03c26723d99891804bb11.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "bf478176ffb5a5264eaebd238ee3a303048eda70ace9a1b9909a687e14bc62c0", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7171597480773926, "target_z": 1.7968750015293478, "ce": 6.133810997009277, "genuineness": 4.193545341491699, "identity": 0.04829762876033783, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/1f9a899910112e9b52888c72.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "c75c70d38f163ec4d0f95aa1eba98a0c2c611d6b2a02b517a6977a222ce75925", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.602978467941284, "target_z": 1.7031250474097839, "ce": 5.935244083404541, "genuineness": 4.099404335021973, "identity": 0.06514353305101395, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/7ed0baf3949b803115de9b73.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "1c8933a0502753fb2a631f5c950e3a56fdba849afab5e35c58af7a1c6f439e97", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2128586769104004, "target_z": 1.382812361594018, "ce": 6.117925643920898, "genuineness": 4.268857479095459, "identity": 0.6456897258758545, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/03773887b9a67b7b08674288.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "0f7d2d81a3d0a9010ade597af0463c21a9a94ce79ae8182e7b021fb5393ffcee", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.298494815826416, "target_z": 1.4531249740010863, "ce": 6.165581703186035, "genuineness": 4.281409740447998, "identity": 0.6358155608177185, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/6a301ea1c1139120c4354e6a.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "ae0ec51ec73ba10586e50816be7ddbbe046f3f1f65d9d82919213ab10a0865f3", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.060616970062256, "target_z": 1.2578124227679326, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.7590066194534302, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/70c8dc4f49076127e93ac0cc.mp3" }, { "key": "original-vn_S_TECH_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "De deur staat nog open.", "Ik heb je jas op de stoel gelegd.", "Zeg maar wanneer je klaar bent." ], "text": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "1d54b596cc18ed6a2cc08ad0e419292bb97f369590e7727e3da2d1b9bb8c8163", "task_id": "vn_S_TECH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] De deur staat nog open.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.660068988800049, "target_z": 1.7499999265913024, "ce": 6.046441555023193, "genuineness": 4.256305694580078, "identity": 0.783962070941925, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/acb60f2a50566125521ddf29.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "50dec5d621878efb86940aa39edfe2184bce424119f872186266253fce0b5982", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.097764253616333, "target_z": 2.109375044351088, "ce": 6.1894097328186035, "genuineness": 4.18099308013916, "identity": 0.5967245101928711, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/a32d42421409d7faf2b1d4f8.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "3e45cf20830db8dc75d9362dd035a1e7c3d87b9712f5d6df179fe5f8f366f0ec", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886143684387207, "target_z": 1.7734374640603252, "ce": 6.086154937744141, "genuineness": 4.231201171875, "identity": 0.6196013689041138, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/b222cb29be91789a3b125102.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "13be1df257996142b2ff014c4196d396acc426168a1969f7be4a565d42d875c2", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3175249099731445, "target_z": 1.4687498684760834, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.7707418203353882, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/85df503bbb49438f2ce99567.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "390a0cb88515d5b263d5a6c609d7372b899c94a3fe0563eab7746d7e1eb22b7d", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2119455337524414, "target_z": 2.2031249984706522, "ce": 6.197352409362793, "genuineness": 4.068024158477783, "identity": 0.7036423683166504, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/acaac3c7ddb3f01824820d5b.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "5c9a2493eb8c6e36a904efa81ef61d0cbeaf52c285ae4123195bed4eb2f7b0c6", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.7701586484909058, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/4ec06b273405d2f71dc60b6d.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.5, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "b8ed83ebb040d297fbbd9d07d7e042478719a244646f5590c1fb03850c989cb0", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8123109340667725, "target_z": 1.8750000611739146, "ce": 6.054384231567383, "genuineness": 4.256305694580078, "identity": 0.6515082120895386, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/787093ac1f63c99f995abcc4.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.75, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "f642784488e1198e5925a44bab7cdfd72e0e31ed0d25809eb6b89e0c53e33269", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1738851070404053, "target_z": 2.1718750137641307, "ce": 6.221180438995361, "genuineness": 3.9927117824554443, "identity": 0.6689813137054443, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/c330617ce92bd66562f1a577.mp3" }, { "key": "original-vn_S_TECH_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_TECH_reg", "language": "nl", "sentences": [ "Ik heb het kleine doosje gevonden.", "Het stond naast het oude raam.", "We kunnen er samen in kijken." ], "text": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "directions": [ "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 1.0, "frames_generated": 128, "seconds_generated": 10.24, "ended": true, "contract_sha256": "028233d337284204aa307f4e123130b9f781f438e2719147d86365105d674df0", "mp3_sha256": "66e5ccdcdfc1b50554d6b6f146693b7e63298249f4dafc635d46f18a902344c3", "task_id": "vn_S_TECH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.8 seconds duration] Het stond naast het oude raam.\n(Strong teacher-like, didactic delivery. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] We kunnen er samen in kijken.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4507365226745605, "target_z": 1.5781249128271715, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.5432798862457275, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/vn_S_TECH_reg/nl/adae22e0ea525f8aa85e4b41.mp3" } ], "source_standard_audio_unchanged": true }