{ "complete": true, "report": { "bucket": "vn_S_AUTH_reg", "label": "Strong Authoritative Style", "definition": { "id": "vn_S_AUTH_reg", "category": "talking_style", "head": "vn_S_AUTH_reg", "pole": "high", "label": "Strong Authoritative Style", "floor": 4.0, "canonical_dimension_name": "Authoritative Style" }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/style_s_auth/full-v2-sealed/sealed-selected", "mission": "style_s_auth", "format_status": "CANONICAL", "heldout": true, "qualified": true, "source_cfg": 3, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.414346151351928, "genuineness": 4.143964409828186, "wer": 0.004430979366463237, "identity": 0.5758873379230499, "voiceclap": 0.02508918127976358, "fitness": 0.8019422827908754, "steering_v2": 0.9448692303230914, "raw_target": 3.116326172351837, "target_cdf": 0.9448692303230914, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 0.99, "by_arm": { "reference": { "n": 50, "ce": 6.41069257736206, "genuineness": 4.104676432609558, "wer": 0.001956989247311828, "identity": 0.5758873379230499, "fitness": 0.7980523600192918, "steering_v2": 0.9367076311370349 }, "none": { "n": 50, "ce": 6.417999725341797, "genuineness": 4.183252387046814, "wer": 0.006904969485614647, "identity": null, "fitness": 0.805832205562459, "steering_v2": 0.953030829509148 } } } } ], "selected_wording": { "id": "manual-553924db97fc", "general": "Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.", "inline": [ "Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.", "Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.", "Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence." ], "cfg": 3, "recipe": "P09", "strategy": "initial-physical-directing", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-3", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.150000028687052, "target_raw": 3.4450739860534667, "ce": 6.177892851829529, "genuineness": 4.209235179424286, "identity": 0.5265093751251697, "wer": 0.013157894736842105, "valid_fraction": 0.95, "identity_delta": -0.027180279046297072, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 2.272265644856819, "target_raw": 3.5720929384231566, "ce": 6.155454802513122, "genuineness": 4.172206556797027, "identity": 0.523458881676197, "wer": 0.034210526315789476, "valid_fraction": 0.95, "identity_delta": -0.030230772495269776, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 2.150000028687052, "target_raw": 3.4450739860534667, "ce": 6.177892851829529, "genuineness": 4.209235179424286, "identity": 0.5265093751251697, "wer": 0.013157894736842105, "valid_fraction": 0.95, "identity_delta": -0.027180279046297072, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 2.052734421123401, "target_raw": 3.344026970863342, "ce": 6.149696326255798, "genuineness": 4.147416245937348, "identity": 0.5822292849421501, "wer": 0.010526315789473684, "valid_fraction": 1, "identity_delta": 0.028539630770683288, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 1.8636718875954088, "target_raw": 3.147614228725433, "ce": 6.193182468414307, "genuineness": 4.258188366889954, "identity": 0.5677826300263404, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.014092975854873657, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 1.8433593777342347, "target_raw": 3.1265120267868043, "ce": 6.169553017616272, "genuineness": 4.272937071323395, "identity": 0.5880102947354316, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.034320640563964847, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": 1.7396484895176938, "target_raw": 3.018769156932831, "ce": 6.20728075504303, "genuineness": 4.276702630519867, "identity": 0.5913993149995804, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.037709660828113556, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": 1.704296890105526, "target_raw": 2.9820431888103487, "ce": 6.223960375785827, "genuineness": 4.337266421318054, "identity": 0.6054876372218132, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.05179798305034637, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": 1.692773444694175, "target_raw": 2.9700717449188234, "ce": 6.203309369087219, "genuineness": 4.324086761474609, "identity": 0.5943600058555603, "wer": 0.029093567251461987, "valid_fraction": 0.95, "identity_delta": 0.04067035168409348, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": 1.6320312624609383, "target_raw": 2.906968080997467, "ce": 6.214032006263733, "genuineness": 4.332559418678284, "identity": 0.5692511923611164, "wer": 0.003125, "valid_fraction": 1, "identity_delta": 0.015561538189649582, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 2.272265644856819, "target_raw": 3.5720929384231566, "ce": 6.155454802513122, "genuineness": 4.172206556797027, "identity": 0.523458881676197, "wer": 0.034210526315789476, "valid_fraction": 0.95, "identity_delta": -0.030230772495269776, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 2.2402343364965973, "target_raw": 3.5388163447380068, "ce": 6.204103660583496, "genuineness": 4.1791102409362795, "identity": 0.5441753815859556, "wer": 0.024303405572755416, "valid_fraction": 0.95, "identity_delta": -0.009514272585511208, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 2.20351560971518, "target_raw": 3.5006700992584228, "ce": 6.110777258872986, "genuineness": 4.13235375881195, "identity": 0.5719527214765548, "wer": 0.03215277777777778, "valid_fraction": 0.9, "identity_delta": 0.018263067305088043, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 2.127734366618002, "target_raw": 3.4219426989555357, "ce": 6.165184569358826, "genuineness": 4.16090966463089, "identity": 0.536189341545105, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.01750031262636185, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 2.121484347254242, "target_raw": 3.415449696779251, "ce": 6.150093460083008, "genuineness": 4.150554227828979, "identity": 0.5496376007795334, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.004052053391933441, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 2.0828124969520005, "target_raw": 3.375274395942688, "ce": 6.175311470031739, "genuineness": 4.205783355236053, "identity": 0.5596134692430497, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005923815071582794, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 2.06914058066954, "target_raw": 3.361070954799652, "ce": 6.148902106285095, "genuineness": 4.2032729387283325, "identity": 0.5587657984346152, "wer": 0.005572755417956656, "valid_fraction": 1, "identity_delta": 0.005076144263148308, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 2.050000028687052, "target_raw": 3.3411862730979918, "ce": 6.227931714057922, "genuineness": 4.220845794677734, "identity": 0.5863483898341656, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.032658735662698744, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 2.034765597523183, "target_raw": 3.325359570980072, "ce": 6.174715757369995, "genuineness": 4.129843282699585, "identity": 0.5257359445095062, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.0279537096619606, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": 2.0089843601634154, "target_raw": 3.298576033115387, "ce": 6.176304292678833, "genuineness": 4.159026885032654, "identity": 0.5588697083294392, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.005180054157972336, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 1.9886718388274203, "target_raw": 3.277473819255829, "ce": 6.201720881462097, "genuineness": 4.178796339035034, "identity": 0.5485018953680992, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.005187758803367614, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 1.96210936697659, "target_raw": 3.2498786747455597, "ce": 6.165581727027893, "genuineness": 4.221787309646606, "identity": 0.5354155674576759, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.018274086713790893, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 1.928515583896833, "target_raw": 3.2149788618087767, "ce": 6.229520201683044, "genuineness": 4.272936940193176, "identity": 0.5938486082479357, "wer": 0.011374269005847953, "valid_fraction": 1, "identity_delta": 0.040158954076468946, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 1.9253906574074042, "target_raw": 3.2117324471473694, "ce": 6.160617518424988, "genuineness": 4.2186492562294005, "identity": 0.570586159825325, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.016896505653858186, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": 1.8886718962015245, "target_raw": 3.173586165904999, "ce": 6.192983937263489, "genuineness": 4.166558170318604, "identity": 0.5653679370880127, "wer": 0.02222222222222222, "valid_fraction": 0.95, "identity_delta": 0.011678282916545869, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": 1.8789062269607113, "target_raw": 3.163440835475922, "ce": 6.199735188484192, "genuineness": 4.277957880496979, "identity": 0.5425522267818451, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.011137427389621734, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": 1.8162109498937031, "target_raw": 3.0983081459999084, "ce": 6.178091430664063, "genuineness": 4.22053210735321, "identity": 0.5451656490564346, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.008524005115032197, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": 1.7955078256557029, "target_raw": 3.0768001437187196, "ce": 6.189012575149536, "genuineness": 4.282351064682007, "identity": 0.5670364677906037, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.01334681361913681, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": 1.7832031342784684, "target_raw": 3.064017081260681, "ce": 6.18424699306488, "genuineness": 4.250657057762146, "identity": 0.569287720322609, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.01559806615114212, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": 1.7542968901055258, "target_raw": 3.033987045288086, "ce": 6.173524379730225, "genuineness": 4.249088084697723, "identity": 0.5512517362833023, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.00243791788816452, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": 1.7289062671225843, "target_raw": 3.0076093077659607, "ce": 6.2219746828079225, "genuineness": 4.316555511951447, "identity": 0.5496787693351507, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0040108848363161085, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": 1.7208984256890028, "target_raw": 2.999290144443512, "ce": 6.221180438995361, "genuineness": 4.292078948020935, "identity": 0.540328423678875, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.013361230492591858, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": 1.675195292844887, "target_raw": 2.9518102049827575, "ce": 6.200529456138611, "genuineness": 4.267602443695068, "identity": 0.5481942117214202, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.00549544245004654, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 2.282031233773886, "target_raw": 3.582238185405731, "ce": 6.201323747634888, "genuineness": 4.152123248577118, "identity": 0.4943880306556821, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.05930162351578474, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": 1.340429695321704, "target_raw": 2.604029881954193, "ce": 6.175311493873596, "genuineness": 4.37617781162262, "identity": 0.5536896541714669, "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": 1.5408203211733509, "target_raw": 2.8122111201286315, "ce": 6.183452701568603, "genuineness": 4.310593223571777, "identity": 0.513001948595047, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.04068770557641983, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": 1.5153320646771393, "target_raw": 2.7857319533824922, "ce": 6.178091382980346, "genuineness": 4.301806795597076, "identity": 0.5779309883713722, "wer": 0.007894736842105263, "valid_fraction": 0.95, "identity_delta": 0.024241334199905394, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": 1.6478515635533526, "target_raw": 2.923403429985046, "ce": 6.1854383707046505, "genuineness": 4.280468225479126, "identity": 0.5641654342412948, "wer": 0.0061111111111111106, "valid_fraction": 1, "identity_delta": 0.010475780069828033, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": 1.555937495732801, "target_raw": 2.827916007041931, "ce": 6.249774055480957, "genuineness": 4.314170608520508, "identity": 0.5820743036270142, "wer": 0.009243459439111612, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 2.308749980062499, "target_raw": 3.6099956798553468, "ce": 6.2729666137695315, "genuineness": 4.106935706138611, "identity": 0.5473741683363914, "wer": 0.005274327122153209, "valid_fraction": 1, "identity_delta": -0.03470013529062271, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 2.3879687283592053, "target_raw": 3.692294225692749, "ce": 6.257081327438354, "genuineness": 4.0835889005661015, "identity": 0.5499366283416748, "wer": 0.0019047619047619048, "valid_fraction": 1, "identity_delta": -0.032137675285339354, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 2.308749980062499, "target_raw": 3.6099956798553468, "ce": 6.2729666137695315, "genuineness": 4.106935706138611, "identity": 0.5473741683363914, "wer": 0.005274327122153209, "valid_fraction": 1, "identity_delta": -0.03470013529062271, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.07921874829670629, "ci95_z": [ -0.017187450263823296, 0.19531244847088255 ], "language_deltas": { "en": 0.12187494441883648, "de": -0.06874992039343022, "fr": 0.3054687045489517, "nl": 0.014062576558570461, "es": 0.023437436350603046 }, "languages_improved": 4, "clear_gain": false, "delta_ce": -0.015885286331176758, "delta_genuineness": -0.023346805572509767, "delta_wer": -0.0033695652173913043, "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": 0.07921874829670629, "ci95_z": [ -0.017187450263823296, 0.19531244847088255 ], "language_deltas": { "en": 0.12187494441883648, "de": -0.06874992039343022, "fr": 0.3054687045489517, "nl": 0.014062576558570461, "es": 0.023437436350603046 }, "languages_improved": 4, "clear_gain": false, "delta_ce": -0.015885286331176758, "delta_genuineness": -0.023346805572509767, "delta_wer": -0.0033695652173913043, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.7528124843296977, "ci95_z": [ 0.6149999868039561, 0.8942187259208056 ], "language_deltas": { "en": 0.6093749515905995, "de": 0.9464843730725886, "fr": 0.6218750017929408, "nl": 0.9691405806695398, "es": 0.6171875145228202 }, "languages_improved": 5, "clear_gain": true, "delta_ce": 0.02319255828857422, "delta_genuineness": -0.20723490238189698, "delta_wer": -0.003969132316958404, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": false, "scale": 0.0, "cfg": 5.0, "reason": "Use manual-only configuration; no clear qualifying incremental adapter gain" }, "reference_only": true, "updated": 1793025333.8252788, "example": { "text": "Ich schrieb die Nummer auf den Umschlag. Lies sie genau... und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "sentences": [ "Ich schrieb die Nummer auf den Umschlag.", "Lies sie genau... und prüfe die letzten beiden Ziffern.", "Ruf an, wenn du Zeit hast." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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.", "language": "de", "seed": 1701, "scale": 0.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "0035e5955d5411fb0f689d477b8d99726022ec2b1339d4792c5b509af56acb36", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9311308cc7f149b6fc80ddbf9b9c97b6f54161b9b6cee29bbe608af2d1aab044", "generation_seconds_batch": 25.711767189088278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06198787689209, "target_z": 1.7812498673223838, "ce": 6.149696350097656, "genuineness": 4.268857479095459, "identity": 0.7356147170066833, "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_AUTH_reg/0035e5955d5411fb0f689d477b8d99726022ec2b1339d4792c5b509af56acb36.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "00386fed222b4cde42a8e5b3b290032aecf7edc5ec3685db55feaf3bd50ca1c1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b3f7ffde0f6589dd11a68f216b81717ec53df8319685a54dd8f2dc7bc9ce093", "generation_seconds_batch": 18.92815965693444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.348263263702393, "genuineness": 4.105680465698242, "identity": 0.6124070286750793, "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_AUTH_reg/00386fed222b4cde42a8e5b3b290032aecf7edc5ec3685db55feaf3bd50ca1c1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "0052101b61770f19597ad05bc26d975bd83c5870777fafbeba2a5b8c14884069", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "649ac3a4d276490d2df29903e5e9d0361849ed766e48e2db276aeec3f1942d87", "generation_seconds_batch": 26.352658731979318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9483609199523926, "target_z": 1.671875087854097, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.5461478233337402, "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_AUTH_reg/0052101b61770f19597ad05bc26d975bd83c5870777fafbeba2a5b8c14884069.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "006524e7c8e2cb288edbd18a7b57741326569d831d0c3b1a64747c41b420e860", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50f9388817759a5b10b0b10bcd9e0d6f0e03fa878be18c272f893021534ff7fa", "generation_seconds_batch": 20.751006614998914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.1894097328186035, "genuineness": 3.923675537109375, "identity": 0.4304490089416504, "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_AUTH_reg/006524e7c8e2cb288edbd18a7b57741326569d831d0c3b1a64747c41b420e860.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "014b4dc3b7ce811134b541faba8f343b33bd7ea310c7e26f2526479a440b5273", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fff9163284a45a23ca8632ab71b318e4d10a8f361f2a53a6a9bcaad53cd153a8", "generation_seconds_batch": 24.690683279070072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.006728172302246, "genuineness": 4.017816066741943, "identity": 0.528727650642395, "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_AUTH_reg/014b4dc3b7ce811134b541faba8f343b33bd7ea310c7e26f2526479a440b5273.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "8540de6bd55c4af2932d4fc36fbde845815de61ae33d7725c93bd5bf2dfa4880", "generation_seconds_batch": 18.076764025026932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6480603218078613, "target_z": 1.3828124166282547, "ce": 6.173524379730225, "genuineness": 4.632867813110352, "identity": 0.7530877590179443, "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_AUTH_reg/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.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": "023a9cb4685c23fee39520e4551ed8b327c5a0a9f259cef03acc7529e2616024", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86d091163b571be5dbabc7eeca28494107e5168bac7632619412dee53a85cdbd", "generation_seconds_batch": 25.918430355959572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.197352409362793, "genuineness": 3.973883628845215, "identity": 0.7040042281150818, "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_AUTH_reg/023a9cb4685c23fee39520e4551ed8b327c5a0a9f259cef03acc7529e2616024.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": "028fc41cbe15b2257ec813dbeae2109c90a6f9d749ebc2dcc05c8fcf984bf273", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe69fb936ae70e0842f263234bbdaca36ebd0aeb41cdeee19096ab006e33534f", "generation_seconds_batch": 22.285085260984488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.300607204437256, "genuineness": 4.118232727050781, "identity": 0.5736188888549805, "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_AUTH_reg/028fc41cbe15b2257ec813dbeae2109c90a6f9d749ebc2dcc05c8fcf984bf273.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "02976df59b1af95aa5c49efafccf15ded386f8bdeeeceacbeacf308570b8c6fb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98a9e6e9d5f80a5dc3053c1787e124f80ace472872e5c175696c8681cab6c827", "generation_seconds_batch": 26.578145529027097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.157639026641846, "genuineness": 4.293961524963379, "identity": 0.8698184490203857, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar 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_AUTH_reg/02976df59b1af95aa5c49efafccf15ded386f8bdeeeceacbeacf308570b8c6fb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "02e711b99241926ecbaf74dc3848d4d3c00711657a05e5bd92c2a9a76451863a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ae0e4e3b5412d5a2cb0b3dd839ab9f6f1458b98dfca2fd70515ec28ed55d7ca", "generation_seconds_batch": 20.443757848930545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7535715103149414, "target_z": 1.4843751380564385, "ce": 6.165581703186035, "genuineness": 4.13078498840332, "identity": 0.7817482948303223, "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_AUTH_reg/02e711b99241926ecbaf74dc3848d4d3c00711657a05e5bd92c2a9a76451863a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "02e8acd7673c81f08ad2df3e22488057ad4b461d9cc15d1651b5c001b507c43b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ce9fe2d55a86731afda16e83f6f652be50684901bd00e462968b06ccc36627c", "generation_seconds_batch": 24.814471708028577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.09409761428833, "genuineness": 4.432034492492676, "identity": 0.8301176428794861, "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_AUTH_reg/02e8acd7673c81f08ad2df3e22488057ad4b461d9cc15d1651b5c001b507c43b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "0371454980d8ae9e08282fcdb300906fb0c48501959f23f921f07dad342de46d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea75be83db1d628068d812f0be833a917c38fcd0971758d514634d6877c9d38b", "generation_seconds_batch": 26.72364959702827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.372091293334961, "genuineness": 3.9801597595214844, "identity": 0.5779900550842285, "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_AUTH_reg/0371454980d8ae9e08282fcdb300906fb0c48501959f23f921f07dad342de46d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "0387491ccace8b4249ddf49a7654ffbb2fb09e9c97d10267e574979b650e9ac0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75cc2d8d9c48eb051ac3e50793096fd17982ead0806fa9bfe00f8ef7d3a4699d", "generation_seconds_batch": 17.296098595950752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.1894097328186035, "genuineness": 4.231201171875, "identity": 0.7958131432533264, "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_AUTH_reg/0387491ccace8b4249ddf49a7654ffbb2fb09e9c97d10267e574979b650e9ac0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "03aa2a73469520bb9bc2d19b80d39128dc6fa9987f1357b07243311d98016007", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1825e433e81a15eb6022d0d7f91920d03e80f37cf53f8ca8d42b553f9031801b", "generation_seconds_batch": 26.936580862966366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.1020402908325195, "genuineness": 4.15588903427124, "identity": 0.43323659896850586, "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_AUTH_reg/03aa2a73469520bb9bc2d19b80d39128dc6fa9987f1357b07243311d98016007.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "03fc0430a283a6ebe97fcf7c5ec3c046b6b1ba3771be745f0c3e462d05cfbdb4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e887d916596ad8628e6b281ad790d0588fd6a7707bc47786c02fc7dac4904ed0", "generation_seconds_batch": 21.310535597032867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.260893821716309, "genuineness": 4.162164688110352, "identity": 0.7053016424179077, "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_AUTH_reg/03fc0430a283a6ebe97fcf7c5ec3c046b6b1ba3771be745f0c3e462d05cfbdb4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "0411780952d6120f830f761988f805b5d94d66213ab7aea071cb9e4dc64ca145", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2ba49dc8486b82e6709fe9e37138dfeb7c961f4a23cdb4a38b048294139062a", "generation_seconds_batch": 24.81232617504429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9622392654418945, "target_z": 0.7226563338199805, "ce": 6.395919322967529, "genuineness": 4.557555198669434, "identity": 0.3485450744628906, "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_AUTH_reg/0411780952d6120f830f761988f805b5d94d66213ab7aea071cb9e4dc64ca145.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "0454d64d7adb8269a8f1a97f71cd828cf9d7b743fab56a39dfad21299e443819", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f26c2545292dafdb127696f1da230e2e932f3e7c7c4b455cc3c52120f5f7a728", "generation_seconds_batch": 24.9501206909772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.623711585998535, "target_z": 1.359374865529443, "ce": 6.459460258483887, "genuineness": 4.369274139404297, "identity": 0.4506387710571289, "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_AUTH_reg/0454d64d7adb8269a8f1a97f71cd828cf9d7b743fab56a39dfad21299e443819.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "04673f00962aeabc2c408f71826de2d87c6554e6bed255bbfed37231c34447f4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "70cbef0ee37380c91f5d99d6d2adda2ca7b6ea972fb02609a39b00c4d76a1633", "generation_seconds_batch": 25.451868592062965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 5.927301406860352, "genuineness": 3.9048471450805664, "identity": 0.6452566385269165, "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_AUTH_reg/04673f00962aeabc2c408f71826de2d87c6554e6bed255bbfed37231c34447f4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "04708c797f2f992fc5d120d95dc2d622c2b8304e3b155b510eca2c4969c4a2d0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68f3c263e4c89ee162afb6ad305371758060f117c7c0188f000902e8326e4ce1", "generation_seconds_batch": 18.400722349062562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4208059310913086, "target_z": 1.1640623986988472, "ce": 6.1894097328186035, "genuineness": 4.519899368286133, "identity": 0.7517467141151428, "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_AUTH_reg/04708c797f2f992fc5d120d95dc2d622c2b8304e3b155b510eca2c4969c4a2d0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "04a85d2ef819d185e6142b0c252fff9e14cb9dd7a4a9f2e4971180a10bb04420", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57be6365c515b96b16692cf89dc58095d9c8ade62de2d4f6c542c5faa188c8a4", "generation_seconds_batch": 18.562955867033452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.263189792633057, "target_z": 2.9375000932329196, "ce": 6.09409761428833, "genuineness": 4.030367851257324, "identity": 0.6082054376602173, "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_AUTH_reg/04a85d2ef819d185e6142b0c252fff9e14cb9dd7a4a9f2e4971180a10bb04420.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "04b5a860ea60558b4e75485b48b6f09c446f9c18d9db5ececf69e1c74c62c975", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6721966e8a70e73ab69aad85654185d63c8f1d41bb77ba7065b66c6c5b3fc5d9", "generation_seconds_batch": 25.081363205914386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.181467056274414, "genuineness": 4.281409740447998, "identity": 0.5188380479812622, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay de aquí.", "wer": 0.11764705882352941, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/04b5a860ea60558b4e75485b48b6f09c446f9c18d9db5ececf69e1c74c62c975.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "0505d9863c39c0b36d54b4c43bfa9a8d1c2ab11678a5786a5a09adb948c16929", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2e0be997d13463f135a4051f44a766c671737b3c343cfbdcaf21ff4079e6423", "generation_seconds_batch": 24.97611093102023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8573784828186035, "target_z": 2.5468751595717274, "ce": 6.23706579208374, "genuineness": 4.231201171875, "identity": 0.585588812828064, "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_AUTH_reg/0505d9863c39c0b36d54b4c43bfa9a8d1c2ab11678a5786a5a09adb948c16929.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "052c2c6b0470e27cd363e7a2e0e55b0c108d3d5bbdf9e63f0ff286d1cd08b6e6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "22801e6867f4b8dc50deb325b00bdeff80b4ddc7d39705bc1cfe7b5af8b3d0b3", "generation_seconds_batch": 26.0075959690148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.786036252975464, "target_z": 1.5156249766917702, "ce": 6.046441555023193, "genuineness": 4.394378662109375, "identity": 0.5734611749649048, "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_AUTH_reg/052c2c6b0470e27cd363e7a2e0e55b0c108d3d5bbdf9e63f0ff286d1cd08b6e6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "0592cb60c33259ecd83f49c14c71997e1169b2dba90855b3fd02100df5445332", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d23fdfc85d6f82c97c6551f3c51f42177f9b6042fa35ccda462cae81d853bd26", "generation_seconds_batch": 24.556657046079636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.100865364074707, "target_z": 2.7812502115670097, "ce": 6.109982967376709, "genuineness": 3.9550557136535645, "identity": 0.6018123030662537, "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_AUTH_reg/0592cb60c33259ecd83f49c14c71997e1169b2dba90855b3fd02100df5445332.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "ed6761d2e484f980ba7d9c9ad0366f6b229a23c861a6f897527cbf056fadbd2f", "generation_seconds_batch": 16.30944906908553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.558781862258911, "target_z": 1.2968749587623625, "ce": 6.372091293334961, "genuineness": 4.595211505889893, "identity": 0.649970293045044, "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_AUTH_reg/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.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": "05995a1fe180a01d55183ff0cdd5a1ed31b3f19481269d882cfb17611820a7b0", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4e6c0b8fc1a4fccb1dfd4ee8844f59d837a9acecaec04469696b9bf403ab408b", "generation_seconds_batch": 22.016084103961475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281126022338867, "target_z": 1.9921873682188542, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.4925457537174225, "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": 0.03999999999999915, "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_AUTH_reg/05995a1fe180a01d55183ff0cdd5a1ed31b3f19481269d882cfb17611820a7b0.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": "05bfd3f4987e5bd1a8f77659dd63a26591c459fbb50ca881b8f07f79aca4c3a6", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f77066a7ac773dbae0bac20642dcf753b1d5304a61b92154bd3c8e03e0d92bcb", "generation_seconds_batch": 22.146845887997188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.356205940246582, "genuineness": 3.9801597595214844, "identity": 0.5203311443328857, "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_AUTH_reg/05bfd3f4987e5bd1a8f77659dd63a26591c459fbb50ca881b8f07f79aca4c3a6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "05f4e02669230221cb028108643219c45bbd3925870fbb6b3ebd471e3d1ccf72", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a165daa86ae1fac245d9ae4969a26e76b7ea2bd3f2c5c39131b85f017ce941ca", "generation_seconds_batch": 25.609792027040385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 6.221180438995361, "genuineness": 4.243753433227539, "identity": 0.5832318067550659, "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_AUTH_reg/05f4e02669230221cb028108643219c45bbd3925870fbb6b3ebd471e3d1ccf72.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "0626b5df46b83f5fb1cbb72bbe2c39e9396a776f36df5fe03f803bb72c0371ea", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57b784c4a74856afe05b38dcd3a6534517561f06e37b879432d866d58797fab7", "generation_seconds_batch": 25.84204577410128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.3820745348930359, "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_AUTH_reg/0626b5df46b83f5fb1cbb72bbe2c39e9396a776f36df5fe03f803bb72c0371ea.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "06567ffff9c7c3986cf4f19382f82ed9f7186a35137b38a20bfbc0d9969e3c8c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fefe6cb7c93ca32f83cacbd8b1a178f6852b5fde41c128417f3d752999c8b693", "generation_seconds_batch": 26.287261795951054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9483609199523926, "target_z": 1.671875087854097, "ce": 6.300607204437256, "genuineness": 4.344170093536377, "identity": 0.6519127488136292, "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_AUTH_reg/06567ffff9c7c3986cf4f19382f82ed9f7186a35137b38a20bfbc0d9969e3c8c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "065adcfd42cf384530bb3bc42c7ae1e9e8408c3352946e38d7bc48e5185dfdb6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c659e9a4c7dd46511a75d4296180753300a8022e76c8034fd01e7328cd532bba", "generation_seconds_batch": 24.923642256995663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0701041221618652, "target_z": 1.789062384355321, "ce": 6.141753673553467, "genuineness": 4.243753433227539, "identity": 0.7213624119758606, "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_AUTH_reg/065adcfd42cf384530bb3bc42c7ae1e9e8408c3352946e38d7bc48e5185dfdb6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "0704958c5b25b4a6121cda69e2ee2ae51363f9d6d21d76df07c8855b800144df", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "273c9c02c3afbf7b1f5bcb03a74ce0e8f767185663e79e39627d44f3ba9b59ea", "generation_seconds_batch": 25.90413449902553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0944528579711914, "target_z": 1.8124999354541327, "ce": 6.403861999511719, "genuineness": 4.469690799713135, "identity": 0.5799344182014465, "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_AUTH_reg/0704958c5b25b4a6121cda69e2ee2ae51363f9d6d21d76df07c8855b800144df.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "07243845a4c091439c7c2df8d1b91374ffcb35af4750ebcdda05e3e8d52ef0e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d7fdc9976a4e02f8a726fde2e11793314eb3df07f34e99fcc75a0b6b6dfd76a6", "generation_seconds_batch": 26.145695246988907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.213237762451172, "genuineness": 4.017816066741943, "identity": 0.36111462116241455, "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_AUTH_reg/07243845a4c091439c7c2df8d1b91374ffcb35af4750ebcdda05e3e8d52ef0e5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "076120d4d712b7997344a9f0e2a797902ce857149857a62e05a1cb5af4a6fbf4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "019ecd890151df6aa6a4f415b5f8a09b4448fb083ea7a0379c755a6caa9e8860", "generation_seconds_batch": 19.347340421983972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.102569103240967, "target_z": 1.82031245248707, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.6207488775253296, "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_AUTH_reg/076120d4d712b7997344a9f0e2a797902ce857149857a62e05a1cb5af4a6fbf4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "08512e785686858028b75c193f56e0cd797d77a3656ac2a8ae5d195bffb6a4dc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd1075f43c65529ad637fd52c175c997e293405284f1af7bb2fa07a28b2ce09f", "generation_seconds_batch": 24.58940027898643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 6.197352409362793, "genuineness": 3.860914945602417, "identity": 0.34157878160476685, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "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_AUTH_reg/08512e785686858028b75c193f56e0cd797d77a3656ac2a8ae5d195bffb6a4dc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "08b80ef37923fc161fada2cba08664fc79e53a7cabc721a8b0a4535b43343ea2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82f9e5af0389d6f2305c58c773454436e21bce2c434b7a250fa312bae2ae00c5", "generation_seconds_batch": 24.65907000703737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932128429412842, "target_z": 1.6562500537882228, "ce": 5.855817794799805, "genuineness": 3.936227321624756, "identity": 0.5520147681236267, "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_AUTH_reg/08b80ef37923fc161fada2cba08664fc79e53a7cabc721a8b0a4535b43343ea2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "094fc2ec752daa780e905b6bd2ef8abe916c975774115ba5c1ad7ffa91a6548e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "10a9542abca90d70166e6006e83db02c7ca7876ece3a1b2b3de211b1fb59ee88", "generation_seconds_batch": 28.0878754099831, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.110685348510742, "target_z": 1.828124969520007, "ce": 6.1894097328186035, "genuineness": 4.570107460021973, "identity": 0.37474143505096436, "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_AUTH_reg/094fc2ec752daa780e905b6bd2ef8abe916c975774115ba5c1ad7ffa91a6548e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "0999db9cff815d2dc6686c9d09abcdb3a6dce90704f011a7996e985dd57ff2f9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "266fe7b42e6a03ee4314df9e0e64fabbe0ebf5c4c967663504a1ed5b14757040", "generation_seconds_batch": 24.37323153205216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.1894097328186035, "genuineness": 4.394378662109375, "identity": 0.5359365940093994, "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_AUTH_reg/0999db9cff815d2dc6686c9d09abcdb3a6dce90704f011a7996e985dd57ff2f9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "09a33362975931d7699a40aaaa30dc48c2d3a74a26fd93a2605c7a847ebed1a4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa8d776859869357c4b59b02f6cf330be1a69a676f7291d5d2fd043ab210715a", "generation_seconds_batch": 25.411230052937753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.980825901031494, "target_z": 1.703125155985846, "ce": 6.165581703186035, "genuineness": 4.457139015197754, "identity": 0.5923655033111572, "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_AUTH_reg/09a33362975931d7699a40aaaa30dc48c2d3a74a26fd93a2605c7a847ebed1a4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "0b2ea9a62e665f71366c9c6e9674a1c45a6fc8e603b31ad3b0d3a7e12ac0dd82", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "37167677e4f85672c385441c360b3c0d2edb1a4b47079c46d9d1d83e4b08f343", "generation_seconds_batch": 25.472863759030588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7941524982452393, "target_z": 1.5234374937247073, "ce": 6.268836498260498, "genuineness": 3.9676074981689453, "identity": 0.6069355010986328, "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_AUTH_reg/0b2ea9a62e665f71366c9c6e9674a1c45a6fc8e603b31ad3b0d3a7e12ac0dd82.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "0b84642d0cb96cbc7232a53a6a8a96829eeaed8a9790ef952b4231a2f68b312c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d96ebcc027380a549fccaa41165f29f2dfd9309a45dd7cb5302d0d9a31902d67", "generation_seconds_batch": 25.046680395957083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.062326908111572, "genuineness": 4.168440818786621, "identity": 0.4972262978553772, "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_AUTH_reg/0b84642d0cb96cbc7232a53a6a8a96829eeaed8a9790ef952b4231a2f68b312c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "0c723efba38da6bb73e53228146ba8c1fdb26244f9e67473308cb275cb642bfd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db359d96705cbc3946411b85045c3637b9f56e57b9f2af200b6c58d21415370f", "generation_seconds_batch": 24.931762705906294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.7063599824905396, "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_AUTH_reg/0c723efba38da6bb73e53228146ba8c1fdb26244f9e67473308cb275cb642bfd.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "f9eb9ac2b19c167e8fc6e9012e6bdb9846d7f83f462f590d6ecfee2135c9b4e0", "generation_seconds_batch": 20.0404176939046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1350340843200684, "target_z": 1.8515625206188187, "ce": 6.117925643920898, "genuineness": 4.545003414154053, "identity": 0.6459794044494629, "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_AUTH_reg/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "0d2c24e81225f8be8848540bdf92c64ddfd517af25c4bca35f3593b322dc13f8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef761cced4bee861325e12ad9ba69393bbb67ec62947de7870045ceeeac80a9d", "generation_seconds_batch": 25.554971315083094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.078212261199951, "genuineness": 4.507347106933594, "identity": 0.42973852157592773, "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_AUTH_reg/0d2c24e81225f8be8848540bdf92c64ddfd517af25c4bca35f3593b322dc13f8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "0d43e01d3375c8a966b99c05afcd58974c8e77f3b610da74d00ccdc6120c00a5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6b2d48bbdf3c77cdcc7f2ee507c2833de24f5b2a8ec0809d0ba9f4a6a0a2e1ef", "generation_seconds_batch": 19.359179602004588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2909464836120605, "target_z": 1.039062585164686, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.7562960386276245, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vommenin. Nous pouvons regarder ensemble ce qu'elle contient.", "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_AUTH_reg/0d43e01d3375c8a966b99c05afcd58974c8e77f3b610da74d00ccdc6120c00a5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "0d56e0cda7494f07f452cfc7ca28c95a3a99605e4dba602fe082dffd1fc71dc2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb7cd470cbea89d5846abc245cb4167b6130c84523aa68bd7fc4b1db8516dd0f", "generation_seconds_batch": 25.142816429026425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7779200077056885, "target_z": 1.507812459658833, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.8090087175369263, "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_AUTH_reg/0d56e0cda7494f07f452cfc7ca28c95a3a99605e4dba602fe082dffd1fc71dc2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "0d62e98f6221d67bde84183799049e5f3d3cb6edc126680aa3bfb6c60a3c1781", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d45f9de9e84647674bcba1338ca31a9bf27ef4135466df1a4f100252bafa5b6", "generation_seconds_batch": 23.81994315504562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.221180438995361, "genuineness": 4.268857479095459, "identity": 0.6020020246505737, "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_AUTH_reg/0d62e98f6221d67bde84183799049e5f3d3cb6edc126680aa3bfb6c60a3c1781.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "0da393c5d40d7d373f3cecfd5280488dc49407a9bcbe813ebdd5067b0b41b2a4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ccd625de529b94d54c007b4d4562b80103df1658b2920c692234df8b21da88e", "generation_seconds_batch": 25.711767189088278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5019683837890625, "target_z": 1.2421875690282191, "ce": 6.09409761428833, "genuineness": 4.344170093536377, "identity": 0.5747465491294861, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Ici, c'est pas le lait, ici, na. Dis-moi quand tu es prêt. Hum.", "wer": 0.4444444444444444, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/0da393c5d40d7d373f3cecfd5280488dc49407a9bcbe813ebdd5067b0b41b2a4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "0dae2553de7b6e2a9f415993e4364bc435932756e12ca777b132d8b9c8962042", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0f9f8b49c41e585546db3db6befc6f48556c824e1f3e36bd520129cc511dcde", "generation_seconds_batch": 24.085979157942347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4092817306518555, "target_z": 3.078124940832955, "ce": 6.221180438995361, "genuineness": 4.080576419830322, "identity": 0.6073439717292786, "hypothesis": "Ik heb het kleine doosje doorgevonden. Het stond naast het oude raam. We kunnen er 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_AUTH_reg/0dae2553de7b6e2a9f415993e4364bc435932756e12ca777b132d8b9c8962042.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "0db5cb6c53aa06ce60ee740bcd974464a13024912eda8ba82e1dc848ca694e8d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e57fa7f99a18e7e08516ee3911168007ac16d4c51842c0dc6e41083be1d5dd68", "generation_seconds_batch": 20.49468479701318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5263171195983887, "target_z": 1.2656251201270308, "ce": 6.427689552307129, "genuineness": 4.413206577301025, "identity": 0.3178124725818634, "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_AUTH_reg/0db5cb6c53aa06ce60ee740bcd974464a13024912eda8ba82e1dc848ca694e8d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "0dd097995abfa89a3a60d06265afed06413c0ebe24bd8ec23942c8fb3e676e0d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dffb54eb2a302f193234b3f911d89a68c5ba4189cf0785db27f238c7d8229430", "generation_seconds_batch": 25.361834149924107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.6688392758369446, "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_AUTH_reg/0dd097995abfa89a3a60d06265afed06413c0ebe24bd8ec23942c8fb3e676e0d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "0e6f8fea4df8c0b8e3a23adbe5d8b00319bfa177221ab779c472719d2c908305", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4ad27620e90777ff2da7b757d3f13c29fc25dd42621e9b3a0a9e0c7d5f7109d", "generation_seconds_batch": 24.37323153205216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4451546669006348, "target_z": 1.1874999497976588, "ce": 6.157639026641846, "genuineness": 4.595211505889893, "identity": 0.7659465074539185, "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_AUTH_reg/0e6f8fea4df8c0b8e3a23adbe5d8b00319bfa177221ab779c472719d2c908305.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "0e743aacc10bbf86e83167922fb8087f856f3c3ea8cf969e34fda572edf984c9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1732b042965f2a5fab47d0ea0fdc62a04c7d48a9962272af37d39a9efa9605a1", "generation_seconds_batch": 26.380035863956437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13332986831665, "target_z": 2.812499820705924, "ce": 6.149696350097656, "genuineness": 4.168440818786621, "identity": 0.636117696762085, "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_AUTH_reg/0e743aacc10bbf86e83167922fb8087f856f3c3ea8cf969e34fda572edf984c9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "0e74df32df55f6c08637cf0c9f64da6ddc28a786800cd9e7f50f16198fcaf293", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f59f5f91b8f77ba799adc64a3598a0bde035bf45466ac9c4cb598372c793e07", "generation_seconds_batch": 26.287261795951054, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.292664527893066, "genuineness": 4.432034492492676, "identity": 0.5227283239364624, "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_AUTH_reg/0e74df32df55f6c08637cf0c9f64da6ddc28a786800cd9e7f50f16198fcaf293.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "0ea0db38698f3349161e269ddd9fbc1069717f03eb7ea09c09a2d45a684464ef", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "ba62865d71c8074c7ca852eb7dc4c043da1e74f1f55ec86f18a60f1af4e8c5a0", "generation_seconds_batch": 20.564030607929453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.09409761428833, "genuineness": 4.18726921081543, "identity": 0.6955162286758423, "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_AUTH_reg/0ea0db38698f3349161e269ddd9fbc1069717f03eb7ea09c09a2d45a684464ef.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "0f5ce4e0f8244263604f9f6b9518612d3f1303dd678c08ce252e2fb4852dd77e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79d55074cd78140c4b022e12809c416eb3ef1b4e0e17dde69b0b7f888126fbcb", "generation_seconds_batch": 25.886080123949796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.834733486175537, "target_z": 1.5624998493929763, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.8419919013977051, "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_AUTH_reg/0f5ce4e0f8244263604f9f6b9518612d3f1303dd678c08ce252e2fb4852dd77e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "0fef7f9069afdff4b9743251d2ca08c3f6f9aee3a7a4e61ee8db4d94a4db9e82", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "202e6d51f2fdf169f330f75e0b91f06874bfec0d4ebb5c3092fd1b290a8a4226", "generation_seconds_batch": 25.72777048300486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.372091293334961, "genuineness": 4.557555198669434, "identity": 0.5029520988464355, "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_AUTH_reg/0fef7f9069afdff4b9743251d2ca08c3f6f9aee3a7a4e61ee8db4d94a4db9e82.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "105eeed1afbbbc192ea933ec75b324efcc70b293ddca79e4e181c9a096cd32b0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71497b4c89e9d80f126b84588fc0bfd24866b56425b249f9863f48fe124ea09c", "generation_seconds_batch": 20.168883408070542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0944528579711914, "target_z": 1.8124999354541327, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.6492328643798828, "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_AUTH_reg/105eeed1afbbbc192ea933ec75b324efcc70b293ddca79e4e181c9a096cd32b0.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": "109d812826128cffbde1d607d0fb44de505074a44df64f72368e91c14b6e38cb", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c5dd58be7942ed2ab33094366399db99e1908154b6b8099d203e02311413183", "generation_seconds_batch": 22.128475266043097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 6.372091293334961, "genuineness": 4.15588903427124, "identity": 0.4798097014427185, "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_AUTH_reg/109d812826128cffbde1d607d0fb44de505074a44df64f72368e91c14b6e38cb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "10dfb65137ea944e2c3dc3509a228dcd51416e48dfffe6dcb054c245f91172f7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8da6f5899440ae2bf30636b615faf0ffeacc708857ba51703c038804148bb0ab", "generation_seconds_batch": 29.309814302017912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0132906436920166, "target_z": 1.7343749946211777, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.5223238468170166, "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_AUTH_reg/10dfb65137ea944e2c3dc3509a228dcd51416e48dfffe6dcb054c245f91172f7.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": "10f7761da784c3f2de4566e3a1fd81e1bce72a23c46c6979bc95a84250736f1f", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cade2338265e0eace3a33d2de7ef580f59a1c1dd8939145ee3a032d7fe4766bc", "generation_seconds_batch": 22.435941111063585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 6.332377910614014, "genuineness": 3.9299514293670654, "identity": 0.3918476402759552, "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_AUTH_reg/10f7761da784c3f2de4566e3a1fd81e1bce72a23c46c6979bc95a84250736f1f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "111a70f90e85dc0e4f4ce4f6b32f59ee0ee577f02f3f617e00b51d6b76dd7130", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a9ce8ce68877351b6dff366f7fcf7e5cce1e4fccb1c0f91c6113d4365e3415e", "generation_seconds_batch": 19.151382499025203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 5.927301406860352, "genuineness": 4.030367851257324, "identity": 0.6429241895675659, "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_AUTH_reg/111a70f90e85dc0e4f4ce4f6b32f59ee0ee577f02f3f617e00b51d6b76dd7130.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "118707a2e3a719b042d776273bd285b8443630918219262a345d6331dae8b740", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "febc98201dab2da079f645baf10bd2d33b6ad119eb47b3f505fc5ffb964328d2", "generation_seconds_batch": 25.234663205919787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.181467056274414, "genuineness": 4.124508857727051, "identity": 0.8006547689437866, "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_AUTH_reg/118707a2e3a719b042d776273bd285b8443630918219262a345d6331dae8b740.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "11d2a0b3eae0ee74d5d4a35bd8a6d3b36cc19c1ee9b16bb0cfa328c71327ed8e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20863abfc978f6cbb5591838226a11595f409a1b1f393d26317ddc0aae6a8170", "generation_seconds_batch": 25.72777048300486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.324435234069824, "genuineness": 4.469690799713135, "identity": 0.7538595199584961, "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_AUTH_reg/11d2a0b3eae0ee74d5d4a35bd8a6d3b36cc19c1ee9b16bb0cfa328c71327ed8e.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "d90a3a8c37d1ed8616c92fefee67769f68834d48edbc2280fdd8ab0317babddb", "generation_seconds_batch": 16.36132884491235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.284721851348877, "genuineness": 4.369274139404297, "identity": 0.5376323461532593, "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_AUTH_reg/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "12b6a559e5541d611b28ee089bf34323439e4f1a8e97aa8b7c18a817ecb24676", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9cc573681d9b003f1069e5ad711f7b142a75cc68a8dc7c52f6748a4372f74d10", "generation_seconds_batch": 25.36220547207631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4857358932495117, "target_z": 1.2265625349623448, "ce": 6.292664527893066, "genuineness": 4.168440818786621, "identity": 0.48954126238822937, "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_AUTH_reg/12b6a559e5541d611b28ee089bf34323439e4f1a8e97aa8b7c18a817ecb24676.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "12efa30048e9838ee96ba14fc3647aacc5799731c909399aa33f170ef85c81ed", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f2fae9e897a5a2cd879187fa9e477ab21e9d7243d51486e42ca6bc16a48536f", "generation_seconds_batch": 25.886080123949796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.09409761428833, "genuineness": 4.074300289154053, "identity": 0.6539547443389893, "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_AUTH_reg/12efa30048e9838ee96ba14fc3647aacc5799731c909399aa33f170ef85c81ed.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "136daeb89c2c6a63b6a703034da68c35fd7d7c9c920cdc7026a5e24a76f8dade", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a438f557b2d52989e4c094283f2600ca05b70463c0bd74f211e7c4a5b810526", "generation_seconds_batch": 26.482058531953953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281126022338867, "target_z": 1.9921873682188542, "ce": 6.022613525390625, "genuineness": 4.13078498840332, "identity": 0.774656355381012, "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_AUTH_reg/136daeb89c2c6a63b6a703034da68c35fd7d7c9c920cdc7026a5e24a76f8dade.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "1425bd4ad3cc78b9b12ab1fe4b9f2e79737196f6bd16ec0e880994d93440619d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a389eb6c18d027e6b794f22af28ed0feeb11cfb8807edbbf00c6cdbf46b84de", "generation_seconds_batch": 19.540063317050226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.149696350097656, "genuineness": 4.3881025314331055, "identity": 0.7063822150230408, "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_AUTH_reg/1425bd4ad3cc78b9b12ab1fe4b9f2e79737196f6bd16ec0e880994d93440619d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "145f6cb5f68619d68f89e8ecd13321c8b11cf0890dce236196503d08268053b7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ccb431996f4d298c89705a1c39ce92be6688ef2eb8b883bf97e54b79ca25fdd", "generation_seconds_batch": 25.13241223699879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.6290030479431152, "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_AUTH_reg/145f6cb5f68619d68f89e8ecd13321c8b11cf0890dce236196503d08268053b7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "14bfd4409817527a9ca7ee3b3bf0ec946f0d88c2c5cb0e671515339c506a33b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b2b0a10037468521dfd63e6f6366c4ffba043975d130c2d5160678629a96f9d", "generation_seconds_batch": 18.682038645027205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664292812347412, "target_z": 1.398437450694129, "ce": 6.23706579208374, "genuineness": 4.595211505889893, "identity": 0.5301167368888855, "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_AUTH_reg/14bfd4409817527a9ca7ee3b3bf0ec946f0d88c2c5cb0e671515339c506a33b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "14e8b73ff11436b37a61da8784011e993ae6ad4d98c711650efbfaf8b3cc5129", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0feef2cd48aee1b7e9598ee27fddddbb7b4a414e7c085a17052a5b082f184517", "generation_seconds_batch": 26.915509214042686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.623711585998535, "target_z": 1.359374865529443, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.6936887502670288, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay detrás.", "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_AUTH_reg/14e8b73ff11436b37a61da8784011e993ae6ad4d98c711650efbfaf8b3cc5129.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "151cedb0c85231dad867348786f656237a998945eb2a01bd98f3f12a76cbb9c8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e558fa989edd80c40ed7d54715b7fc4bb44cae2a2f0bdfe043eedf93d44bb9f5", "generation_seconds_batch": 24.62843315396458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.157639026641846, "genuineness": 4.143336772918701, "identity": 0.6940271258354187, "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_AUTH_reg/151cedb0c85231dad867348786f656237a998945eb2a01bd98f3f12a76cbb9c8.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "e4cd1ddcf3131507e960fa55d6009bfe8db7cdc3c8ab3da939e28b7423c01260", "generation_seconds_batch": 19.793420445057563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.81038498878479, "target_z": 1.5390625277905818, "ce": 6.2767791748046875, "genuineness": 4.281409740447998, "identity": 0.6331416964530945, "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_AUTH_reg/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "16f074cb1511bdaa6a1622d61deca5d6a17a4928a2334893f2964d72d8e96ab6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9ddd0bdaf57436f2170a77cc3061694132c7802ed589eccddb81b39b1a08b85f", "generation_seconds_batch": 26.32413364900276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.05963397026062, "target_z": 0.8164063087188098, "ce": 6.260893821716309, "genuineness": 4.300237655639648, "identity": 0.4108402132987976, "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_AUTH_reg/16f074cb1511bdaa6a1622d61deca5d6a17a4928a2334893f2964d72d8e96ab6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "17152bb2d09e3b5d30c5fb231f8f68b5e0e9db20a3de81416b059eee715c3735", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "5c29015120ce957e82910ba99dd44724760107a404547ebd063983ab84f01074", "generation_seconds_batch": 20.150925190071575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.534433364868164, "target_z": 1.273437637159968, "ce": 6.133810997009277, "genuineness": 4.356721878051758, "identity": 0.7082736492156982, "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_AUTH_reg/17152bb2d09e3b5d30c5fb231f8f68b5e0e9db20a3de81416b059eee715c3735.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": "173b948694740d5a565e5d222eba058aee5273d04268f1f50ae8036be3522cb5", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a60e1052f2c44b5ed4cb1c6c89e3d9ffc521f2753bb3595e4ce1b4f2b9e64b6b", "generation_seconds_batch": 22.143539529992267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.165581703186035, "genuineness": 4.306513786315918, "identity": 0.7624495029449463, "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_AUTH_reg/173b948694740d5a565e5d222eba058aee5273d04268f1f50ae8036be3522cb5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "1776bec7b0a3015154b6f1841f52277b40eada6076e33d6bd7cf4eb9191accc4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b46c246a561302a498608fcdef5c53a4a21d1f21eec273119c81b6aa6ba370d", "generation_seconds_batch": 24.66350451495964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.587838649749756, "target_z": 3.2499998565647394, "ce": 6.197352409362793, "genuineness": 4.168440818786621, "identity": 0.6165810823440552, "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_AUTH_reg/1776bec7b0a3015154b6f1841f52277b40eada6076e33d6bd7cf4eb9191accc4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "177bb15aff47203285245161bd1f2c9def4c72d31742ff0aba2b5e37a6e82277", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d042c3dd8fd51d7db9f052b5d7c0b64324c463eece867fe72d837d58ab0b26e6", "generation_seconds_batch": 20.302960409899242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1999640464782715, "target_z": 1.9140626568823165, "ce": 6.221180438995361, "genuineness": 4.325342178344727, "identity": 0.5917879939079285, "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_AUTH_reg/177bb15aff47203285245161bd1f2c9def4c72d31742ff0aba2b5e37a6e82277.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "18426a54d9d049682a26c15f7a20fd59a74370b42f8f9d71374b9bfbe4e27bce", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98c1b7bc149526f25182561ac9b3ec1baf8d8abf59c6fb5bdc7a8300ead26ad1", "generation_seconds_batch": 26.280841688043438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 6.205295085906982, "genuineness": 4.055472373962402, "identity": 0.6490261554718018, "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_AUTH_reg/18426a54d9d049682a26c15f7a20fd59a74370b42f8f9d71374b9bfbe4e27bce.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "18961123aa5b1fa631dd87aedb2a29d3d300714a629558889df7d46300d996fd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6800f16207d325ee39e7978c65918b388ff7911b064c7c73604e3ef7adcb2c08", "generation_seconds_batch": 24.14635851001367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.229123115539551, "genuineness": 3.7793264389038086, "identity": 0.5434223413467407, "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_AUTH_reg/18961123aa5b1fa631dd87aedb2a29d3d300714a629558889df7d46300d996fd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "18cb474a85cec50a01ccf63ef5bd7447746fab45dd7b83af0068cb3f940e0ad8", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "038dbb6603ad69d0f189c9d32f75d1f0944b03f9e8d7f82d379a0e67f58079c3", "generation_seconds_batch": 20.1104445370147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 6.2767791748046875, "genuineness": 4.519899368286133, "identity": 0.737790584564209, "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_AUTH_reg/18cb474a85cec50a01ccf63ef5bd7447746fab45dd7b83af0068cb3f940e0ad8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "18d3607a491d481ccdcd58641b2e4fddef0f86dc39a73e7ccbf8bdf6eaecfe3f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c83344f499c8f86904bab19368c93a6ebca9c6fdf3ac26e22a466d55c2c44e18", "generation_seconds_batch": 24.578243616968393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.086154937744141, "genuineness": 4.344170093536377, "identity": 0.766371488571167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Quand tu es prêt, tu es prêt.", "wer": 0.3888888888888889, "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_AUTH_reg/18d3607a491d481ccdcd58641b2e4fddef0f86dc39a73e7ccbf8bdf6eaecfe3f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "19029e8cc11ca09e160a3a15e4864ef1d04f04b98462b19aca534ea99e27c848", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e07a4560335aae1aae6e8cc9c583ec56ff5136990b56a8cd6048450ac787167", "generation_seconds_batch": 24.350154315005057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.3641486167907715, "genuineness": 4.519899368286133, "identity": 0.6602101922035217, "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_AUTH_reg/19029e8cc11ca09e160a3a15e4864ef1d04f04b98462b19aca534ea99e27c848.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "192eafa493257e4e4ab14689b87a7853ebfb82efdce49549954272e9549f5d17", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "328bf4c6e979a6e3e3219ff720058d9c66400fa2de507f79baa0de61fed1d834", "generation_seconds_batch": 20.063804444973357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 5.637393951416016, "genuineness": 4.400654315948486, "identity": 0.6407755613327026, "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_AUTH_reg/192eafa493257e4e4ab14689b87a7853ebfb82efdce49549954272e9549f5d17.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "197f0a83b5f826992d31513822321af6cf04e76d3a3c6670faef6674145be3bd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "044ab26f542b71d350fc99ee81f8b2d0ed72517f8b7c8c9190521b8296e1b873", "generation_seconds_batch": 26.515055862953886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.5811421871185303, "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_AUTH_reg/197f0a83b5f826992d31513822321af6cf04e76d3a3c6670faef6674145be3bd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "1997657b8dbd039f2339ca89db349f693b99ab7d0efe69a7c54b11de6048b5f1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0787e94bdb7b39b5279a047181c92974922d6761d173204a4d0309e3c1fa18c", "generation_seconds_batch": 19.99374872399494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.221180438995361, "genuineness": 4.419482707977295, "identity": 0.7547726035118103, "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_AUTH_reg/1997657b8dbd039f2339ca89db349f693b99ab7d0efe69a7c54b11de6048b5f1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "19a529a42fdcfe3f53abf16691b5168cb4dc54b3874669ab91050ebdf8726607", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "edf65aad3ec87343b318763cc1aa6a380d4d1c51f1e729cd0827e96b6a70e0d7", "generation_seconds_batch": 26.751399541040882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.5626062154769897, "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_AUTH_reg/19a529a42fdcfe3f53abf16691b5168cb4dc54b3874669ab91050ebdf8726607.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": "19e4a76f3feec3423061951e5b5bc4c7cace850656f715fcd4a7ca0aa9c1611d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bffbaaa32b7d9a459dbf7d2e945214767ea15a10d99cd9d56d2e755150598ad0", "generation_seconds_batch": 16.752967583946884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.141753673553467, "genuineness": 3.8546390533447266, "identity": 0.5102790594100952, "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_AUTH_reg/19e4a76f3feec3423061951e5b5bc4c7cace850656f715fcd4a7ca0aa9c1611d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "1a25b201fe3525308de4cb15275941fb4dd2d7eec90912035758b5e601572fc1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5781e416ab010bb621defc8ea2277f2c6793c28602b2568436f58ed55d1cae5", "generation_seconds_batch": 27.998524766997434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.165581703186035, "genuineness": 4.287685871124268, "identity": 0.6884561777114868, "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_AUTH_reg/1a25b201fe3525308de4cb15275941fb4dd2d7eec90912035758b5e601572fc1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "1b04165bb1930870f1711a873082182337205d19142d415e45ab8c2f36526698", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "075e5e39adb876ff39fe5b0162e5a4d5d317b1d7cd3806c89591e5a57c01ad31", "generation_seconds_batch": 19.844501759973355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1350340843200684, "target_z": 1.8515625206188187, "ce": 6.070269584655762, "genuineness": 4.256305694580078, "identity": 0.5724582076072693, "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_AUTH_reg/1b04165bb1930870f1711a873082182337205d19142d415e45ab8c2f36526698.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "1b706b3cdbf01bd91836e269908d9ea9816afecf3cfce43460a1aae1e5c77445", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d03bbed45f89e3321acc6995e7de968be320ebe8c770671fb743fd08733cb59", "generation_seconds_batch": 24.14635851001367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.308549880981445, "genuineness": 4.105680465698242, "identity": 0.7168717384338379, "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_AUTH_reg/1b706b3cdbf01bd91836e269908d9ea9816afecf3cfce43460a1aae1e5c77445.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "1c5217ccfe0698759b3b8eea471e2b394ee0e26db95271e774489bf8e5e963b3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5d818e354d85f7ba36d1842be4c6b46a0659feb35aec682d450e0b739791400e", "generation_seconds_batch": 19.48725551192183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4353346824645996, "target_z": 2.140625191844661, "ce": 6.173524379730225, "genuineness": 3.8546390533447266, "identity": 0.35953569412231445, "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_AUTH_reg/1c5217ccfe0698759b3b8eea471e2b394ee0e26db95271e774489bf8e5e963b3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "1c741f8250754e99e44bf9065eea245a73e6067bdc6bfb0b77486dd9137dac26", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "081086655c8c86ee271bdb53bba055c445ef64bea1edf28113f1dda4bf816baf", "generation_seconds_batch": 21.399306531995535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.110685348510742, "target_z": 1.828124969520007, "ce": 6.070269584655762, "genuineness": 4.344170093536377, "identity": 0.7068721055984497, "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_AUTH_reg/1c741f8250754e99e44bf9065eea245a73e6067bdc6bfb0b77486dd9137dac26.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": "1d2f7100f8b2d977250002c78e1188a38644c77b9c609c727098b8b62498b0a0", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20a1d8abe5d5c435767e688c44aeaa39c534d6e7fbd9772ae9c5f9dd3aa83b41", "generation_seconds_batch": 16.469661337905563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.300607204437256, "genuineness": 4.250029563903809, "identity": 0.6137146949768066, "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_AUTH_reg/1d2f7100f8b2d977250002c78e1188a38644c77b9c609c727098b8b62498b0a0.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "04e79ae7b2149f47947f3efc871f133b4972950aa96b0b921cc337b017c5723b", "generation_seconds_batch": 17.311654831981286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664292812347412, "target_z": 1.398437450694129, "ce": 6.109982967376709, "genuineness": 4.256305694580078, "identity": 0.7192074060440063, "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_AUTH_reg/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "1d7b9431c90049ccd77a1e2136597b1010a6bc9157289609467e5d73adce8dd9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4bbde5bd00b0f33a5b0efbb2a1d6470b88271f4596a46ab08a669a16a67c7d0d", "generation_seconds_batch": 24.976513386005536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.50667667388916, "target_z": 3.1718751452282015, "ce": 6.23706579208374, "genuineness": 4.162164688110352, "identity": 0.3450900912284851, "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_AUTH_reg/1d7b9431c90049ccd77a1e2136597b1010a6bc9157289609467e5d73adce8dd9.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "2fd93d8c0578abde93a4e166af86a795ef7b5161ac367a6615d8cf83ae9555d8", "generation_seconds_batch": 17.86702163692098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.282830238342285, "target_z": 1.031250068131749, "ce": 6.284721851348877, "genuineness": 4.457139015197754, "identity": 0.6025524139404297, "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_AUTH_reg/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "1da199411b29ad5c277510cf01dbb52392534c25bb594df5fce094e783d38cef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ef18c0f38fd9330b7e0fe4c395696542e0906abd373eb49be4df6df10812dc3", "generation_seconds_batch": 29.353176041971892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 5.9431867599487305, "genuineness": 3.973883628845215, "identity": 0.45474886894226074, "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_AUTH_reg/1da199411b29ad5c277510cf01dbb52392534c25bb594df5fce094e783d38cef.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "1dfde5f0937378297a0ba975daa7d1f3fa16d58b7a5f6914b9e9e83d7c383abc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ed13e92764c502eb63393ab711562d71dde04edc41781b59ac51f9373f0ce49", "generation_seconds_batch": 25.382786588044837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 6.197352409362793, "genuineness": 4.068024158477783, "identity": 0.7160304188728333, "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_AUTH_reg/1dfde5f0937378297a0ba975daa7d1f3fa16d58b7a5f6914b9e9e83d7c383abc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "1e4612416a536ebe462d8a867a9318329ec9b295ba9dd2e24e7ede2dd9b4757e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ccab9598834b0c9ccc8adaa8cdc90ff37890a104f2b7c5d34ed50dff1d2af31a", "generation_seconds_batch": 20.250342253013514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1042730808258057, "target_z": 0.8593749229035473, "ce": 6.395919322967529, "genuineness": 4.356721878051758, "identity": 0.36296346783638, "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_AUTH_reg/1e4612416a536ebe462d8a867a9318329ec9b295ba9dd2e24e7ede2dd9b4757e.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": "1e49348233813c993a412cf56766cf87971158027ea9acb9a4ac9f92da4b1863", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "fa9b008018ae41cbfd439bd37c52bfc58d1519b801665b427ee7f1b2dd1373cb", "generation_seconds_batch": 25.697462563053705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.284721851348877, "genuineness": 4.344170093536377, "identity": 0.8175250291824341, "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": 0.120000000000001, "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_AUTH_reg/1e49348233813c993a412cf56766cf87971158027ea9acb9a4ac9f92da4b1863.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "1e99ad1fb28c689b94a6ba84e5c58e898926bc0393afba5b18920fef3cdc25f0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "669fcf3c546e3a532e0c179ee2bbe086509a6906448479d315f7fe0074901f53", "generation_seconds_batch": 26.223303742939606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.229123115539551, "genuineness": 4.18099308013916, "identity": 0.626773476600647, "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_AUTH_reg/1e99ad1fb28c689b94a6ba84e5c58e898926bc0393afba5b18920fef3cdc25f0.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "457b3b858d54519822c60a15fc9c8f23e71109c44286bd16f90a9f5e11d7e8fb", "generation_seconds_batch": 17.311654831981286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281126022338867, "target_z": 1.9921873682188542, "ce": 6.23706579208374, "genuineness": 4.042920112609863, "identity": 0.722724437713623, "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_AUTH_reg/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "1eb1b35a1e9b672540ccc880bdc5d0ccb074bcc6367dceaeaddbd26f5693e88b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a82c913ded2cb5056fe7a7b2a2983479746d7d5085b39623bff0a2459ca5bcbf", "generation_seconds_batch": 24.2462028299924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.149696350097656, "genuineness": 4.262581825256348, "identity": 0.3352304995059967, "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_AUTH_reg/1eb1b35a1e9b672540ccc880bdc5d0ccb074bcc6367dceaeaddbd26f5693e88b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "1eb973f8b5d4b4017c4cf9c54ef1df7ba907a71122d7f80ba02a87c36ffb3eda", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d299eb724f95f540d32951fa5ba901c252696244d63d8c4edf8e4f27a025c736", "generation_seconds_batch": 25.586016373010352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.348263263702393, "genuineness": 4.15588903427124, "identity": 0.5756853818893433, "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_AUTH_reg/1eb973f8b5d4b4017c4cf9c54ef1df7ba907a71122d7f80ba02a87c36ffb3eda.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "1f35f191b0dd559a1bdac59a365f75435401fa5ddb946cf4d1976389c1f7a975", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e9935bc0baeccc4b548cd5f77d7dbabaa65e5b00ce80a3cd612b83cadf8d0dd", "generation_seconds_batch": 25.442441481980495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.332377910614014, "genuineness": 4.055472373962402, "identity": 0.5429337024688721, "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_AUTH_reg/1f35f191b0dd559a1bdac59a365f75435401fa5ddb946cf4d1976389c1f7a975.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "1f6293598c43025b376d8c0f0e36676e381df7c6cae25b699248dd67b7f66979", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdfd18dbf049e994db8cdb072db1dca0e691c6fda9b49ab9f3ba18a429e96fed", "generation_seconds_batch": 25.669120797072537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5425496101379395, "target_z": 1.2812501541929053, "ce": 6.324435234069824, "genuineness": 4.143336772918701, "identity": 0.2629900574684143, "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_AUTH_reg/1f6293598c43025b376d8c0f0e36676e381df7c6cae25b699248dd67b7f66979.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "1f96236c91d00e107954d496697b305b5e6d37bee2255749aa4e6fbb2d8b28a5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f05002c3bd529fd026ca4d809c1c80598db4ce2e7d567bb324bf3e15484700fc", "generation_seconds_batch": 24.734309680992737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8671984672546387, "target_z": 1.5937499175247252, "ce": 6.197352409362793, "genuineness": 4.381826400756836, "identity": 0.47806239128112793, "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_AUTH_reg/1f96236c91d00e107954d496697b305b5e6d37bee2255749aa4e6fbb2d8b28a5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "204fce34ec11c44d5699e65655e590e270492bb472ae6fb97aa5e35d92d41908", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "241a690bca9510032a2b96c6507219200ec91c3902dd4efa5b9076baa62498e8", "generation_seconds_batch": 25.617274954915047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.558781862258911, "target_z": 1.2968749587623625, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.8532477021217346, "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_AUTH_reg/204fce34ec11c44d5699e65655e590e270492bb472ae6fb97aa5e35d92d41908.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "206608ae331cd5f6a901ff8000546193602749b4fca9f40405bf23bbf05664b6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "75f8c5d1f64da2a47c201f75bb86e4b6b70d61d8e8223eaf8af9451c3f7361a3", "generation_seconds_batch": 20.705880695022643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3558762073516846, "target_z": 1.1015624919317666, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.8175888061523438, "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_AUTH_reg/206608ae331cd5f6a901ff8000546193602749b4fca9f40405bf23bbf05664b6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "20c90c0520cbfef7b1ed6dcb55d24721722e70691c0844838128adc458803ef0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdda73727fa2294fd3ca4641d4a33ce66af6444fd8070ee23b46d668dd59aaa4", "generation_seconds_batch": 25.960712919011712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.81038498878479, "target_z": 1.5390625277905818, "ce": 6.332377910614014, "genuineness": 4.281409740447998, "identity": 0.7189871072769165, "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_AUTH_reg/20c90c0520cbfef7b1ed6dcb55d24721722e70691c0844838128adc458803ef0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "210118360040fcf073e2ee8e488c9dd58e438175a93a7cc02b72d48d2c2a91e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5eacbf4a91bbe133b6e2eb8507b9c5d5584a8937bc273c6e5533a8168d8ab7e", "generation_seconds_batch": 26.86438907601405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.372091293334961, "genuineness": 4.1998209953308105, "identity": 0.5214382410049438, "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_AUTH_reg/210118360040fcf073e2ee8e488c9dd58e438175a93a7cc02b72d48d2c2a91e5.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": "2120fd36a291b5e9a1cdbaef26525389a099feaa20c1ce00c407195b49794c85", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "051ea2f8f601892f3a83460317be5ace71936faa39d7f008b35bfc924e64d932", "generation_seconds_batch": 22.155357246054336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.344351768493652, "target_z": 3.015624804569457, "ce": 6.125868320465088, "genuineness": 4.15588903427124, "identity": 0.39360082149505615, "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_AUTH_reg/2120fd36a291b5e9a1cdbaef26525389a099feaa20c1ce00c407195b49794c85.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "2241d292a39dfd36ed32f2bdd091a798760fcfcb55aca08b4b94411b1f49c63f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ff82408a9ea6dc161f94a1a665ab818fbc8e62dee8a08e152b3a908ecf9f35d", "generation_seconds_batch": 25.13280539191328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.197352409362793, "genuineness": 4.344170093536377, "identity": 0.47720521688461304, "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_AUTH_reg/2241d292a39dfd36ed32f2bdd091a798760fcfcb55aca08b4b94411b1f49c63f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "2253c34a80bd48076d00edc8a2dfae892972ac494f08eb9be1f6566297177372", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f417b1ca9148373470c19f4ed9fbd9679ccf9e06f33b29c0ccb1346aa4487b8e", "generation_seconds_batch": 25.962707411963493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.205295085906982, "genuineness": 3.973883628845215, "identity": 0.2557508647441864, "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_AUTH_reg/2253c34a80bd48076d00edc8a2dfae892972ac494f08eb9be1f6566297177372.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "2289d36033851fadf9b610a03653955f73820442c9473ed22c70044b8f04ddcf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d773b0c9abcf97367ace1ebdff77ea8adbba599c240d99457046d7198ca6c2a3", "generation_seconds_batch": 27.341825028066523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664292812347412, "target_z": 1.398437450694129, "ce": 6.3641486167907715, "genuineness": 4.331617832183838, "identity": 0.49348685145378113, "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_AUTH_reg/2289d36033851fadf9b610a03653955f73820442c9473ed22c70044b8f04ddcf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "228eeed5a2cdbd81485924f9b4d9cfcfa35a6a9966c479878ba7bda9861ec03c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fcda7405fefb42fb583c0bb7d6acc4a9c2ca937f37d59d5d0ce49a2beeca678", "generation_seconds_batch": 28.93567006511148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7924485206604004, "target_z": 2.48437502330823, "ce": 6.348263263702393, "genuineness": 4.15588903427124, "identity": 0.5202963948249817, "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_AUTH_reg/228eeed5a2cdbd81485924f9b4d9cfcfa35a6a9966c479878ba7bda9861ec03c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "2292be9b73d0ac548020aa74f28c7956d3f2b4caca355e9f9287919f45757a6b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffcf7ed179434a921c7abf4576c3d2d6064285920430f09225152096f4d9e8d2", "generation_seconds_batch": 24.453199386945926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 6.117925643920898, "genuineness": 4.293961524963379, "identity": 0.4527532458305359, "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_AUTH_reg/2292be9b73d0ac548020aa74f28c7956d3f2b4caca355e9f9287919f45757a6b.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": "23168add4e14611b4cbf78c8fb9e28f4e10808de853f540547745a847509b22b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f78a4f5474b2d0bdae2ce06f3e3c28757613235c2cf29738e3daae180228cb1c", "generation_seconds_batch": 16.48313725506887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.292664527893066, "genuineness": 4.118232727050781, "identity": 0.5447796583175659, "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_AUTH_reg/23168add4e14611b4cbf78c8fb9e28f4e10808de853f540547745a847509b22b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "231b6e834461c4cd06501396ce650a08154e9959e0e278c56fb89cd7fcc3c9a4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f7e44f564e1690216f547ca31b06323c848edb41561a560d1b376e7093ed6cfa", "generation_seconds_batch": 29.353176041971892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 5.974957466125488, "genuineness": 4.080576419830322, "identity": 0.7228258848190308, "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_AUTH_reg/231b6e834461c4cd06501396ce650a08154e9959e0e278c56fb89cd7fcc3c9a4.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": "233910e986faf4a0b539b2333228a3f3710656bb2e665d7d2eb282355666397e", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "20492343a5c3e9f8909837cfd1871b28b44144fef21aae7a66c44cd80f9467b2", "generation_seconds_batch": 16.615672801970504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.197352409362793, "genuineness": 4.268857479095459, "identity": 0.6452131271362305, "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": 0.03999999999999915, "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_AUTH_reg/233910e986faf4a0b539b2333228a3f3710656bb2e665d7d2eb282355666397e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "2461787661fa2c30be5ea1b913fe54951f318e0ecec4ba7332d6b78a7a009737", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2434fe51b35cbc1f538ee49a562b63f113d1823ce2a585b021714138b923f5d6", "generation_seconds_batch": 24.418968347017653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9483609199523926, "target_z": 1.671875087854097, "ce": 6.252951145172119, "genuineness": 4.024092197418213, "identity": 0.21890845894813538, "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_AUTH_reg/2461787661fa2c30be5ea1b913fe54951f318e0ecec4ba7332d6b78a7a009737.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "249817d2195c642b3bf4a4b78cbc1004c339e6a9c1b2aa0c34f38be02630d87d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "798b28fda4e95f89848a31650216ffec67e18a6ddd4e641b2bef219dedb3ae20", "generation_seconds_batch": 25.280982254073024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.125868320465088, "genuineness": 4.494794845581055, "identity": 0.6040123701095581, "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_AUTH_reg/249817d2195c642b3bf4a4b78cbc1004c339e6a9c1b2aa0c34f38be02630d87d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "249ed61dee7377b602c723143c633e77ce03d71f9f6365f610baadfa0b87e60d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e81aa9f05cbed3cb4cb188479288e8bb63356a07172f6bdeadae62f25a92b92", "generation_seconds_batch": 18.108147385064512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.300607204437256, "genuineness": 4.425758361816406, "identity": 0.470978319644928, "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_AUTH_reg/249ed61dee7377b602c723143c633e77ce03d71f9f6365f610baadfa0b87e60d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "2550c1c7a2745ef6c48b1b0eaa31176b95f3804fe056c232e36b68a754d35f3a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c96d0b3cb1a36249c263793d52dbdfb1a83f633683ec8f26605ba2cd8cd2fdf", "generation_seconds_batch": 19.99853840400465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8185012340545654, "target_z": 1.546875044823519, "ce": 6.260893821716309, "genuineness": 4.2751336097717285, "identity": 0.5643675327301025, "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_AUTH_reg/2550c1c7a2745ef6c48b1b0eaa31176b95f3804fe056c232e36b68a754d35f3a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "263136c5b82c87339590c5b758817a21dc4e989148d6ef6da908a06cda9b13a0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "71864d8e9ca40738677e7a8c452939f31fd1abd3e5a3236fd6689a9f81a58f90", "generation_seconds_batch": 26.693047498934902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.054384231567383, "genuineness": 4.250029563903809, "identity": 0.6657366752624512, "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_AUTH_reg/263136c5b82c87339590c5b758817a21dc4e989148d6ef6da908a06cda9b13a0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "266c73c3d47c5292260ef98a3dc301c016ca03226930f75883e3400db72ca0ec", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d72728bf6cfbf38539edc4644f78088e6ae1134e057d062a37e57df90c3c6d7", "generation_seconds_batch": 25.188401284976862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.550665855407715, "target_z": 1.2890626712258424, "ce": 5.8875885009765625, "genuineness": 4.262581825256348, "identity": 0.6557421684265137, "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_AUTH_reg/266c73c3d47c5292260ef98a3dc301c016ca03226930f75883e3400db72ca0ec.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "22a4d113ff6b24427a6c26f12df91b987ac8a4b6fee8732604ebcc400948e793", "generation_seconds_batch": 16.221659452887252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4613871574401855, "target_z": 1.2031249838635332, "ce": 6.4515180587768555, "genuineness": 4.268857479095459, "identity": 0.3647860884666443, "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_AUTH_reg/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.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": "276f068559c188bce5b87d76a3ac093e962e2eee2d14cdfdd2c91eb0733a2ff3", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c0a9d3188e9fd3db5d8c7f0848b8f3762d62c4dae810601427a29f69aa899a5", "generation_seconds_batch": 22.130793037940748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.395919322967529, "genuineness": 3.8797430992126465, "identity": 0.629650890827179, "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_AUTH_reg/276f068559c188bce5b87d76a3ac093e962e2eee2d14cdfdd2c91eb0733a2ff3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "27a831d06cf3480f1ad3b1cd69120078d7a4f4fcb54d21be1ea125b12c55dbc1", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "9fa4342995b5dc651151b2dd7cdbe037900952cc84527f2ce7cb00fefac78e5c", "generation_seconds_batch": 20.1104445370147, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 6.197352409362793, "genuineness": 4.570107460021973, "identity": 0.7047173380851746, "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_AUTH_reg/27a831d06cf3480f1ad3b1cd69120078d7a4f4fcb54d21be1ea125b12c55dbc1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "27e24ec723accbc272aaa0f1d8c28b61d3c14b0f9dfb2aa2a19af069bd6b85f3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aae1c0bddf9e9dc6543930ed8a95312b2fd76ae7164d9adf2ce446c5ec86df41", "generation_seconds_batch": 26.548717948026024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.054384231567383, "genuineness": 4.149612903594971, "identity": 0.4269612431526184, "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_AUTH_reg/27e24ec723accbc272aaa0f1d8c28b61d3c14b0f9dfb2aa2a19af069bd6b85f3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "27edaad4e3dc4517bfb2d99a24ce5ba774a14a91b509e783831c3c77937e4d9a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db410d69badd470e2a1f6c88be8bc688f5b912234fb43ebea8f2b76ac17de846", "generation_seconds_batch": 26.373753060004674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3315277099609375, "target_z": 1.0781251703293722, "ce": 6.221180438995361, "genuineness": 4.532451152801514, "identity": 0.7227036356925964, "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_AUTH_reg/27edaad4e3dc4517bfb2d99a24ce5ba774a14a91b509e783831c3c77937e4d9a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "28a58d7d430c09d807c308fb4f7665ef707325092d5f3cb97a17e3b639cb7189", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d181e016a4ac6797c7fe029e19006da46e3492e5015b28989f51618c312948cd", "generation_seconds_batch": 26.751399541040882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.157639026641846, "genuineness": 4.080576419830322, "identity": 0.5337953567504883, "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_AUTH_reg/28a58d7d430c09d807c308fb4f7665ef707325092d5f3cb97a17e3b639cb7189.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "28e8104e092a18fbfed5c59f46200d3e7d598801e40768e345e58cb5919573e4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed7d262dc989bee9fe9e9d98c3613e1f4be40e7a8bcb659cb449be99c9582214", "generation_seconds_batch": 24.179298240924254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7762160301208496, "target_z": 2.4687499892423554, "ce": 5.95112943649292, "genuineness": 3.9299514293670654, "identity": 0.537097692489624, "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_AUTH_reg/28e8104e092a18fbfed5c59f46200d3e7d598801e40768e345e58cb5919573e4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "299f5adefc7d9aecfa4c2e85be96d67d1e650c85126a9a26a5e12bd042232e3b", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "0338eca6ee7240ec084e9ae1225e4089df65e948eee68bd35cedbd39d2d40fb8", "generation_seconds_batch": 20.564030607929453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 6.316492557525635, "genuineness": 4.356721878051758, "identity": 0.6465736031532288, "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_AUTH_reg/299f5adefc7d9aecfa4c2e85be96d67d1e650c85126a9a26a5e12bd042232e3b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "29c405b692c00d4940817f6ae05b295b561d1e8c5ece02ad348c90c92656ddc2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b4efb7d91d6ae28db893cc576e8116a971c466dc9acbf1a2a0f3337ed14034d", "generation_seconds_batch": 24.63029351597652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.295654773712158, "target_z": 2.9687501613646683, "ce": 6.181467056274414, "genuineness": 3.8169827461242676, "identity": 0.5072161555290222, "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_AUTH_reg/29c405b692c00d4940817f6ae05b295b561d1e8c5ece02ad348c90c92656ddc2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "29fd11fcca786074fd66853a4183c4e5ba2cd0b2d387be5419665224c09cdb34", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a9d543a84fbfbdcd04f6cfa99cb64f2c915a10bcda80abca251fcb0faa02c51", "generation_seconds_batch": 25.52629535808228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.308549880981445, "genuineness": 4.519899368286133, "identity": 0.5657374858856201, "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_AUTH_reg/29fd11fcca786074fd66853a4183c4e5ba2cd0b2d387be5419665224c09cdb34.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "2b2be550c22d0c6602b47745e6f1d4c63496d7ebd5d1da017f5e0c7421cc8058", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d6403ba07e9af363a8370eb347228527e5ada1aa7bd8cd9d1dc8da7e6be64bb", "generation_seconds_batch": 25.0076725439867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.7485747337341309, "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_AUTH_reg/2b2be550c22d0c6602b47745e6f1d4c63496d7ebd5d1da017f5e0c7421cc8058.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "2b33d9df4c93b6cb37298ea058c182ed87b45ca478d4f507c59c417204635357", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8b98ff9d69029196a6ca497c30fc2aaf6a7178913d0ade6e04f544dfe7642f1a", "generation_seconds_batch": 26.054093715036288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.260893821716309, "genuineness": 3.9927117824554443, "identity": 0.34515663981437683, "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_AUTH_reg/2b33d9df4c93b6cb37298ea058c182ed87b45ca478d4f507c59c417204635357.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "2b74d374a364d8ab3096a013992daff80d90f530c327601993f5255935a3736f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88fc3c42fdc616e33d407145600d85464865080452a9e5bd9c02d62a3d373bd8", "generation_seconds_batch": 26.794133082963526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.181467056274414, "genuineness": 4.218649387359619, "identity": 0.6006430387496948, "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_AUTH_reg/2b74d374a364d8ab3096a013992daff80d90f530c327601993f5255935a3736f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "2b7e2460ce4fa991d592cc082384568c97c53f92b4e002baeb57f3e7bb37d75a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e16dd72e883f79d549dc5bb2a89f4c530cbb46d7698f042024d357b945582918", "generation_seconds_batch": 26.15496964205522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.086154937744141, "genuineness": 3.823258876800537, "identity": 0.6966781616210938, "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_AUTH_reg/2b7e2460ce4fa991d592cc082384568c97c53f92b4e002baeb57f3e7bb37d75a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "2b901c92c47653b33d4ecb3ee21c079f4e9605ac50db1391616628fbd0275b7d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e6b740b2b7c5af75316af9bff3f7f75e4e17ebc9509bb4501e6b18c8cc3ad76", "generation_seconds_batch": 25.585969454958104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.110685348510742, "target_z": 1.828124969520007, "ce": 6.260893821716309, "genuineness": 4.017816066741943, "identity": 0.40508949756622314, "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_AUTH_reg/2b901c92c47653b33d4ecb3ee21c079f4e9605ac50db1391616628fbd0275b7d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "2b92e54d90badcbfe8667496feb030ae4bc33c948a7d37839874ca42c8bc166f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30052285b4f8e1d0737de461abe1abdc3b3a1aae86d192333b923063a4c8991a", "generation_seconds_batch": 24.645064475014806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 5.990842819213867, "genuineness": 4.036643981933594, "identity": 0.6074425578117371, "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_AUTH_reg/2b92e54d90badcbfe8667496feb030ae4bc33c948a7d37839874ca42c8bc166f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "2bb56a47ed6f300bab11b2a3ce91537f2ceef61d1709ec22077c19bf691c45ed", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efdef67370bbd6e5f79194c547b5af29c92033eb141a2204955582df6d3e346e", "generation_seconds_batch": 25.590263979975134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.599363088607788, "target_z": 1.3359375439270487, "ce": 6.205295085906982, "genuineness": 4.281409740447998, "identity": 0.5880337357521057, "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_AUTH_reg/2bb56a47ed6f300bab11b2a3ce91537f2ceef61d1709ec22077c19bf691c45ed.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": "2be658c0302bde8f1a7b5a5812c27c6e8478e6c99e1301b13deb8812061ebde5", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f9dcac670eb866a0d1a4c21c4bbe72bbb01ca3341db01b766a6e40acbc943f6c", "generation_seconds_batch": 26.15458785602823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0684003829956055, "target_z": 2.7500001434352606, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.608190655708313, "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_AUTH_reg/2be658c0302bde8f1a7b5a5812c27c6e8478e6c99e1301b13deb8812061ebde5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "2c0181d4e8714dd878a1f88c2573f89bf546c0d93c08c731495c2376732a31f8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d3fc79c0b2c1d474add3d1712b329fd73aaeeca56e1c4710db14424b0cd6075", "generation_seconds_batch": 18.518168642069213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.38003396987915, "genuineness": 4.193545341491699, "identity": 0.48097753524780273, "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_AUTH_reg/2c0181d4e8714dd878a1f88c2573f89bf546c0d93c08c731495c2376732a31f8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "2c0ac5560a13c05107a18783842f1b84758f75e0283692356c93fa6a3354c077", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58a5f4c71c49b0f1e406ee942b8c82e9eb619de044d6f1f8f80f99299210b967", "generation_seconds_batch": 25.617274954915047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5831305980682373, "target_z": 1.320312509861174, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.7184269428253174, "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_AUTH_reg/2c0ac5560a13c05107a18783842f1b84758f75e0283692356c93fa6a3354c077.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "2d83735390589f7630aafe8b5bbbc643d2fe45034cfdbd98bee4664ba7a795fb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3962eb78a85cd63025c64568b7d68f51b011e5f85f8bfc26fb8c1bd883750b8d", "generation_seconds_batch": 25.13280539191328, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1982598304748535, "target_z": 2.874999956969422, "ce": 6.09409761428833, "genuineness": 4.005263805389404, "identity": 0.685084342956543, "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_AUTH_reg/2d83735390589f7630aafe8b5bbbc643d2fe45034cfdbd98bee4664ba7a795fb.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": "2d9d292eda3e70e7a9d9c8b727d40e6a91ad3bb00fe6a46f4af7efe543d64044", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b6b5fcc17d9c05c7f7d9c79650c2cbda9c7650f33de372321de43437f6ca88b", "generation_seconds_batch": 19.581194686004892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.443574905395508, "genuineness": 4.030367851257324, "identity": 0.4076744019985199, "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_AUTH_reg/2d9d292eda3e70e7a9d9c8b727d40e6a91ad3bb00fe6a46f4af7efe543d64044.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "2ea562003f7d17fccfe09d981628219afeab4d1332602459872c3e17d115c999", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9501eb100b3723b20151438d125a2a6e863e5f0a25a892b14f822cdb4e26c632", "generation_seconds_batch": 25.763939644908533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.268836498260498, "genuineness": 4.005263805389404, "identity": 0.3119683861732483, "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_AUTH_reg/2ea562003f7d17fccfe09d981628219afeab4d1332602459872c3e17d115c999.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "2eac2c5eb811b032b4bd1d6fc4171af4b4f98a1a1f4664f60d2ee4ae8e1d81e8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "85a79e60951918beb74b4e8ccbdf65ec7228e7b02176cc88890c018f7bdc7031", "generation_seconds_batch": 26.363409465993755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402869701385498, "target_z": 2.1093751237129124, "ce": 6.284721851348877, "genuineness": 3.9927117824554443, "identity": 0.1185583844780922, "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_AUTH_reg/2eac2c5eb811b032b4bd1d6fc4171af4b4f98a1a1f4664f60d2ee4ae8e1d81e8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "2edd96ca88b4371cac2becfe9baa3fa4191f3d8e21ffb2250317c7e19ea99ba6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "91e9657a2c21a2bc2811f488dbad8ced385d9d777917242b4e2eba2830fd99d6", "generation_seconds_batch": 25.01931671402417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664292812347412, "target_z": 1.398437450694129, "ce": 6.038498878479004, "genuineness": 4.400654315948486, "identity": 0.6879898309707642, "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_AUTH_reg/2edd96ca88b4371cac2becfe9baa3fa4191f3d8e21ffb2250317c7e19ea99ba6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "2f0ef5a5a01cd6a0636f7cc382664b4ab820a4013dfa242191df29e53a8d3649", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "050a7651e015ff0826f4328c01c07d18532f39f0ecf8802b2d69f1415e04969e", "generation_seconds_batch": 18.687926208018325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.875314712524414, "target_z": 1.6015624345576622, "ce": 6.070269584655762, "genuineness": 4.293961524963379, "identity": 0.6139683723449707, "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_AUTH_reg/2f0ef5a5a01cd6a0636f7cc382664b4ab820a4013dfa242191df29e53a8d3649.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "2f5ea201c78d26cc4db95b4911d143c1b04b6f0011545338c83b9d105836ae78", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32ed5705c4de08b73341a08c294803ce5a637d6ba1051f2493f982cead1ea1dc", "generation_seconds_batch": 20.03564080502838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.006728172302246, "genuineness": 4.15588903427124, "identity": 0.6311957240104675, "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_AUTH_reg/2f5ea201c78d26cc4db95b4911d143c1b04b6f0011545338c83b9d105836ae78.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "2f795073dde07be732f721f16938b2e04c506639d18a707e16f76572a26a62e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d188368ed28cb2e0486ecaacd602fc41d94c350f02844b9a0e019132099094f9", "generation_seconds_batch": 25.85885436506942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 5.990842819213867, "genuineness": 4.1998209953308105, "identity": 0.6724426746368408, "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_AUTH_reg/2f795073dde07be732f721f16938b2e04c506639d18a707e16f76572a26a62e5.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": "2fdf4141312b89c391ba92e385635b3ebfb4df04a5fb2a2883aad6dea2ad04ec", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "878d3c2f9813111427951a6a77aae934901132aa072bdc9304ce4e1887c9cdec", "generation_seconds_batch": 22.37587234005332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.38797664642334, "genuineness": 3.9173994064331055, "identity": 0.424304336309433, "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_AUTH_reg/2fdf4141312b89c391ba92e385635b3ebfb4df04a5fb2a2883aad6dea2ad04ec.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "30c3938c45a201cbd4929e71052809a3d086db8a202be25cbeef71e56ef060bf", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 128, "seconds_generated": 10.24, "mp3_sha256": "903cda31212391214914ef2ec05e976ed33118cd12ef71cd8be6e859335f74a3", "generation_seconds_batch": 18.304570907028392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.956477165222168, "target_z": 1.6796876048870344, "ce": 6.260893821716309, "genuineness": 4.030367851257324, "identity": 0.16617928445339203, "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_AUTH_reg/30c3938c45a201cbd4929e71052809a3d086db8a202be25cbeef71e56ef060bf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "30ca755b6d7060bd61e54ac39168fa041ef32edd27776540b01f71b3d662eec8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3cdf34c8922626ff90f8f4cf5b61bff36e906df03dd6c2daea60d0b34b3a2d67", "generation_seconds_batch": 25.90413449902553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.597659111022949, "target_z": 2.296875073510571, "ce": 6.197352409362793, "genuineness": 4.344170093536377, "identity": 0.2212890386581421, "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_AUTH_reg/30ca755b6d7060bd61e54ac39168fa041ef32edd27776540b01f71b3d662eec8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "30cb0c076c3e2fa9210195ae463cd22bafebfbeb523505500b50fc6cd0d14339", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "007c1be4fc4525167bda6d5d6e56b7d9bfcb362e67ae2da1be9e0c40c75b10c4", "generation_seconds_batch": 24.256153885042295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5182008743286133, "target_z": 1.2578126030940937, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.700056791305542, "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_AUTH_reg/30cb0c076c3e2fa9210195ae463cd22bafebfbeb523505500b50fc6cd0d14339.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "311e356023a65a18a6b71f353c2aab83c1d01d499ed732393d8d031daae195cd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88916964a06d0da3471caea5b84b635b28d56dc0d307db535cb89fa8b3e0f24b", "generation_seconds_batch": 27.705699809943326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.054384231567383, "genuineness": 4.13078498840332, "identity": 0.4173926115036011, "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_AUTH_reg/311e356023a65a18a6b71f353c2aab83c1d01d499ed732393d8d031daae195cd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "312c2f080d67022b3f4e29b47c8f29b7936b03f4b27a9cd600fbceb21c34a4c3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cc61ab8fa1a17fc8d33f4a8942c88cf36d4bfc7d31e4a75ef3079c2409798792", "generation_seconds_batch": 18.66718391003087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.281126022338867, "target_z": 1.9921873682188542, "ce": 6.205295085906982, "genuineness": 4.074300289154053, "identity": 0.3815590739250183, "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_AUTH_reg/312c2f080d67022b3f4e29b47c8f29b7936b03f4b27a9cd600fbceb21c34a4c3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "3174f7509e8ca926ecabc032dd54679b6efbb63fb4819a0833600fa304d5fdbc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9937d97ad27473e7920dd99efb2c960272d7684eb408df5643f09312581987f", "generation_seconds_batch": 19.502909668022767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4353346824645996, "target_z": 2.140625191844661, "ce": 6.157639026641846, "genuineness": 3.9927117824554443, "identity": 0.3753935396671295, "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_AUTH_reg/3174f7509e8ca926ecabc032dd54679b6efbb63fb4819a0833600fa304d5fdbc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "318860e71d15506357ba8629159b290e7153a01a342c42f0c0ad709020300a44", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cd46dd4f0703f2b65661bf75e433b8c9e6b4a7dbbca15a159bc453f308eef45", "generation_seconds_batch": 26.779353736899793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4451546669006348, "target_z": 1.1874999497976588, "ce": 6.316492557525635, "genuineness": 4.287685871124268, "identity": 0.44853663444519043, "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_AUTH_reg/318860e71d15506357ba8629159b290e7153a01a342c42f0c0ad709020300a44.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "32a142d48bb584b212e1ff6a67b066d2d782f15a03042301102847d74d4129f8", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "1bc7e8ad8cf240a1a0ca080dc780814f2d0e560dba8225ce26033eeab15213d6", "generation_seconds_batch": 20.349048105999827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.300607204437256, "genuineness": 4.337893962860107, "identity": 0.7138941287994385, "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_AUTH_reg/32a142d48bb584b212e1ff6a67b066d2d782f15a03042301102847d74d4129f8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "332cad380a6c208d6d8b8dd5bf58b428d3557e0b615ea05e840bb8564fd6e5b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "375d80be2929f3cf122941ef5e62a638753bba21acb1a8a87cbcf51f108a811b", "generation_seconds_batch": 19.287579694064334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.38003396987915, "genuineness": 4.093128204345703, "identity": 0.580651044845581, "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_AUTH_reg/332cad380a6c208d6d8b8dd5bf58b428d3557e0b615ea05e840bb8564fd6e5b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "336cf3335f61cf36bd70f4fdaad819c0babce4bc3dadaff7e1c75335a40898e6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a040dcfa9e8d0c04a28a40c462684be5c4b95e59b4a0d13014a38eafb65d9da8", "generation_seconds_batch": 25.60245196998585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.052167892456055, "target_z": 2.7343751093693864, "ce": 5.927301406860352, "genuineness": 3.911123275756836, "identity": 0.6422388553619385, "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_AUTH_reg/336cf3335f61cf36bd70f4fdaad819c0babce4bc3dadaff7e1c75335a40898e6.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "c5b98f5da03796f3c26f5d4fe1e361664e6593a5b9bcf429d91a201dd1d3ff04", "generation_seconds_batch": 18.107821139972657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.205726146697998, "target_z": 0.9570313858152625, "ce": 6.181467056274414, "genuineness": 4.419482707977295, "identity": 0.5437242388725281, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "188bfc6bf83ec40925fca8d358a4d1db36b035c3cc86ba0ad33b9021a12ede52", "generation_seconds_batch": 20.615297571988776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.403861999511719, "genuineness": 3.7542223930358887, "identity": 0.27920034527778625, "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_AUTH_reg/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "3559d1970a24d7b444b2d15243a522c7ddd15f1d08e79a020064c45bbdc7683b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efbc12a108aa90f54ff08ad9dd3791f617266529bf77989f104d4b12d362b5bb", "generation_seconds_batch": 25.43126228894107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3071789741516113, "target_z": 1.0546876192305605, "ce": 6.006728172302246, "genuineness": 4.632867813110352, "identity": 0.6983085870742798, "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_AUTH_reg/3559d1970a24d7b444b2d15243a522c7ddd15f1d08e79a020064c45bbdc7683b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "35b715a232d28712b24e38b27234d4531e6186556339c3ba27612c6f1d3b38c0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f930abf1efe88f55a8b989e9137888b7a973f583bb0312d823e0f501a443ffb", "generation_seconds_batch": 24.25232342001982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.284721851348877, "genuineness": 4.099404335021973, "identity": 0.6037471294403076, "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_AUTH_reg/35b715a232d28712b24e38b27234d4531e6186556339c3ba27612c6f1d3b38c0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "35dff127ac0032f81064f35045c94de8b112b4f8d2bb621a6d1088aee2b5a0dc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07c6f76a423f28e3159f1efff9bba412264075c30a8ecd43fb5b24d189a3f89e", "generation_seconds_batch": 28.29972495196853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.0146708488464355, "genuineness": 4.381826400756836, "identity": 0.6957061886787415, "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_AUTH_reg/35dff127ac0032f81064f35045c94de8b112b4f8d2bb621a6d1088aee2b5a0dc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "35ec72323fc46f0546beae2b136befec0048b911106847b912e9d1fc9d8576b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f6c8b7a455d482236c08ccca95e61f0326c300341713726fc53ef3e5f1aa6fd", "generation_seconds_batch": 24.52841027698014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.252951145172119, "genuineness": 4.005263805389404, "identity": 0.607380747795105, "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_AUTH_reg/35ec72323fc46f0546beae2b136befec0048b911106847b912e9d1fc9d8576b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "35f091e183df9b3e2316d1fb9e646b8325e0a15623a5e14eb56f88ea6016cca1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5e4787eeffe8720e90010aa72091f27049a6dc305d1b2cf07b8bc0b67301d36", "generation_seconds_batch": 19.131115519907326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 6.395919322967529, "genuineness": 4.331617832183838, "identity": 0.5917477011680603, "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_AUTH_reg/35f091e183df9b3e2316d1fb9e646b8325e0a15623a5e14eb56f88ea6016cca1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "35fcefdc696f017196aca8a27764d84998fc1c895e3f7d43c974227cd4f8e4fc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ead3f3566ad8cdf60476647251c831b4c01a8b7d68c00c2b6da7d9caba6319ed", "generation_seconds_batch": 19.595160383963957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.425514221191406, "target_z": 3.0937499748988295, "ce": 6.149696350097656, "genuineness": 4.093128204345703, "identity": 0.7752646803855896, "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_AUTH_reg/35fcefdc696f017196aca8a27764d84998fc1c895e3f7d43c974227cd4f8e4fc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "36774e735b565963ef4b5b3b037560f54c19a61bce76808f20b0de8efcaa4a8c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ed8d14e57a37b84e15fd77778cc0144e4bf955eaf840e20bb3e5a887e2ebfc3d", "generation_seconds_batch": 25.16909567802213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.680525302886963, "target_z": 1.4140624847600036, "ce": 6.38797664642334, "genuineness": 4.250029563903809, "identity": 0.38631492853164673, "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_AUTH_reg/36774e735b565963ef4b5b3b037560f54c19a61bce76808f20b0de8efcaa4a8c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "36b9e8401a9baa03542efae65d60ea339ba7606aec2a39a56b706e518dfff82a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b55bb2e819f9d6784de2edab89b33e850ee8c75352663401768ce3203fefddb4", "generation_seconds_batch": 19.916308910935186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.09409761428833, "genuineness": 4.331617832183838, "identity": 0.6040722131729126, "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_AUTH_reg/36b9e8401a9baa03542efae65d60ea339ba7606aec2a39a56b706e518dfff82a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "36f71eae6ac09be17058e56ebe816c7145319ec50067869bffcc990cc1781595", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "6bfdcb70352d8d6f7c5a119059858feee7f4e3e0023336d8c99fb3bccb06ad3e", "generation_seconds_batch": 18.304570907028392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0132906436920166, "target_z": 1.7343749946211777, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.35873541235923767, "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_AUTH_reg/36f71eae6ac09be17058e56ebe816c7145319ec50067869bffcc990cc1781595.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": "3700279d3321834b02a6343abc989842bc3450d0d11145cc23c513af6f251833", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6620d6400ac4d7a25f95136d79a5a24eb74e72e85d56f4982585dbe3e5477437", "generation_seconds_batch": 16.485854727914557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 6.09409761428833, "genuineness": 4.080576419830322, "identity": 0.5978518128395081, "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_AUTH_reg/3700279d3321834b02a6343abc989842bc3450d0d11145cc23c513af6f251833.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "374940d5ebc80bbfaf2d3ca78ebaae0f9e1cbcb177a02a9ef141150d9c504f15", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aafa4be89305d0c72860765c23af61c18897b3f5c70390a445234598bc7f8a6e", "generation_seconds_batch": 18.168829747010022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 5.998785495758057, "genuineness": 3.9676074981689453, "identity": 0.5724844932556152, "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_AUTH_reg/374940d5ebc80bbfaf2d3ca78ebaae0f9e1cbcb177a02a9ef141150d9c504f15.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "377860c753de8b67f529ed38ebf70fa773e2e0d198f05cd138142683882721c7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0801d5c54c35677aef58743dcc149d155a6a9881a5df46a9a6b77de313f41c8d", "generation_seconds_batch": 25.126038472983055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.157639026641846, "genuineness": 4.055472373962402, "identity": 0.6649280786514282, "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_AUTH_reg/377860c753de8b67f529ed38ebf70fa773e2e0d198f05cd138142683882721c7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "378ecac56b8076409b337027ca9f3cdc67074eaf5b512653235b5cb0d8f0634a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "353f270bef4d52361212e5bb7cf03b7f5ded7eeaa36f7ee559aff1edc5cee5b9", "generation_seconds_batch": 25.60245196998585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.441746711730957, "target_z": 3.1093750089647036, "ce": 6.284721851348877, "genuineness": 4.149612903594971, "identity": 0.39638176560401917, "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_AUTH_reg/378ecac56b8076409b337027ca9f3cdc67074eaf5b512653235b5cb0d8f0634a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "3818ee7fc870ec2328294151e8ba96fa137093b0c4803ee8d13b020be6f1d23a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52774502bf9e2fc1136cd47d9b5fad97a1d4b44f9d210ba23393edf66db1c0c7", "generation_seconds_batch": 19.595160383963957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.100865364074707, "target_z": 2.7812502115670097, "ce": 6.205295085906982, "genuineness": 4.055472373962402, "identity": 0.6163431406021118, "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_AUTH_reg/3818ee7fc870ec2328294151e8ba96fa137093b0c4803ee8d13b020be6f1d23a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "382cd64d9cd2f296d2f7b957e15558a166af4a689efb568988f2a584a7dae2ef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d9ccb94c6737e5f70480be3077fbd21054dbf8f9d8be3cba41d9061b7a83266", "generation_seconds_batch": 25.0199925879715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.23706579208374, "genuineness": 4.193545341491699, "identity": 0.5022847652435303, "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_AUTH_reg/382cd64d9cd2f296d2f7b957e15558a166af4a689efb568988f2a584a7dae2ef.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "384e42bec1105bd14fccd33afb694ef7ff24da9a2171e72f8454398f5e32ff3d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f821273b83dcf23bbbf4eac4662b30e590dea9923f0e83892c427cd04c090771", "generation_seconds_batch": 20.82471571408678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.125868320465088, "genuineness": 4.143336772918701, "identity": 0.6344506740570068, "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_AUTH_reg/384e42bec1105bd14fccd33afb694ef7ff24da9a2171e72f8454398f5e32ff3d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "3857b4c2bfe67fb073c5b73728aac01a14057ab20af87179f12d7890759ffc8b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2704763490c24294843379740c5dcf6c42b2ffade95754a172434f1b44376e32", "generation_seconds_batch": 26.382820946979336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.6175585389137268, "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_AUTH_reg/3857b4c2bfe67fb073c5b73728aac01a14057ab20af87179f12d7890759ffc8b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "38ac95a4ae42d6867a3a6dc0c975e2c37a92d53e35d7fd4a06df28f2afa954e3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13ccdc9eca4c8fc3b6d7696393fe44a479d3fe3b1966589e9b8dfd818a90a147", "generation_seconds_batch": 19.611200220067985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.078212261199951, "genuineness": 4.2751336097717285, "identity": 0.688622772693634, "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_AUTH_reg/38ac95a4ae42d6867a3a6dc0c975e2c37a92d53e35d7fd4a06df28f2afa954e3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "38c2ee3bd6256d5b4d058baba05e76d6f63e207929dcff53d49eb5b37e0e8456", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "130bd612bedb9361ac03addbf13b1956be4b6565759884cc9c6f9a42e7167bf5", "generation_seconds_batch": 24.81232617504429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404573440551758, "target_z": 1.1484373646329726, "ce": 6.38797664642334, "genuineness": 4.281409740447998, "identity": 0.35359442234039307, "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_AUTH_reg/38c2ee3bd6256d5b4d058baba05e76d6f63e207929dcff53d49eb5b37e0e8456.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "39265231ee2937be07561a0c05b96f4ccd46d932ee4519902cd0cc6fca251132", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33e4426333549ef21b42ca0ce982297a2470e00a5ac5cdf83d1bfd5888a750f3", "generation_seconds_batch": 24.899244813947007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.110685348510742, "target_z": 1.828124969520007, "ce": 5.967014789581299, "genuineness": 4.193545341491699, "identity": 0.5511802434921265, "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_AUTH_reg/39265231ee2937be07561a0c05b96f4ccd46d932ee4519902cd0cc6fca251132.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "3929d65f9248b298bf37d38ae1fa7ca6e8c47995c70aafac1154ab4ea7fbc7bd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "366e5b53adac5f6309d70d334e8033de4bcd5b2ef408dc800f1a1d71b31b57d4", "generation_seconds_batch": 20.751006614998914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.292664527893066, "genuineness": 3.8797430992126465, "identity": 0.17146286368370056, "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_AUTH_reg/3929d65f9248b298bf37d38ae1fa7ca6e8c47995c70aafac1154ab4ea7fbc7bd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "392db8571f2719bc79b7f01525ec73852e0ee7102124d718bb20590cf099fc2b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "438b303450e1e91108cadf1154f8ee99acb50550c45dd6f88d2e88262b326146", "generation_seconds_batch": 26.7274280450074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.205295085906982, "genuineness": 4.356721878051758, "identity": 0.3865357041358948, "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_AUTH_reg/392db8571f2719bc79b7f01525ec73852e0ee7102124d718bb20590cf099fc2b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "393ff0d578634b392e61d5a45dc96e8a7d669976a4573fd8e8c3cb7ccd4b04cc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ae9b0f68e73fc30316e3169f25001bedadbf9eed659f8e7553b4f274bc8e2cd", "generation_seconds_batch": 24.176364543032832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.356205940246582, "genuineness": 4.036643981933594, "identity": 0.6102318167686462, "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_AUTH_reg/393ff0d578634b392e61d5a45dc96e8a7d669976a4573fd8e8c3cb7ccd4b04cc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "398b1ac52babe09884efbbf6e7be1e01a8add7aae297a4de69dc03c9cb86a067", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e0f51aab34295edfefe3f4ac61ce64f95fbb3f2292fd00ca4cf003665ceb22a0", "generation_seconds_batch": 25.78359319700394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 5.935244083404541, "genuineness": 3.9550557136535645, "identity": 0.6893885135650635, "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_AUTH_reg/398b1ac52babe09884efbbf6e7be1e01a8add7aae297a4de69dc03c9cb86a067.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "399221c394b4628cb864316b355e2cae6778d962dbbaa65a9063a9a9b690bd54", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3211fa6d1a94d89b1468746149b65b12ef8247025ea8258a2586d15ff27f7d70", "generation_seconds_batch": 25.00538860098459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6967577934265137, "target_z": 1.429687518825878, "ce": 6.300607204437256, "genuineness": 4.582659721374512, "identity": 0.7229873538017273, "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_AUTH_reg/399221c394b4628cb864316b355e2cae6778d962dbbaa65a9063a9a9b690bd54.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": "39c15ac8f48c6b3e0dc5aa010761d95c0105bba838fa7109d31dd3269327a62d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4351615a795c9063916fdc48b9f8aba9c404601d7931b42ec6a422c4e9168c47", "generation_seconds_batch": 24.988654232933186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.348263263702393, "genuineness": 3.9425034523010254, "identity": 0.40463119745254517, "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_AUTH_reg/39c15ac8f48c6b3e0dc5aa010761d95c0105bba838fa7109d31dd3269327a62d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "39e6721f2003238263e3f7fc56e587ba3ec594ebcb4b77da92d6e93067f02fdf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c8503fc3c9ad40fe7d935b2292fc45c77d9908234ca3437c57763eafef8dcf2", "generation_seconds_batch": 25.971301213023253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.173524379730225, "genuineness": 3.9425034523010254, "identity": 0.3083434998989105, "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_AUTH_reg/39e6721f2003238263e3f7fc56e587ba3ec594ebcb4b77da92d6e93067f02fdf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "39efd80ee02dc7b0d78f2f25aae5ea6d308dc0d0823c913616d323519d46f509", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a17c188d2b30006ff9012e7e47eb4e7b5e4d3954a598a97fc3a3e7e7f3a35b44", "generation_seconds_batch": 25.89979123591911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.213237762451172, "genuineness": 4.457139015197754, "identity": 0.40368562936782837, "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_AUTH_reg/39efd80ee02dc7b0d78f2f25aae5ea6d308dc0d0823c913616d323519d46f509.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": "3a576102b8dab43090e1abca259c1abcd50c6c0b5192a566d64a65251123adbc", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7f8eed5404a6c7ce257283ab4fd985581185dd463439e20e27384a5791250abf", "generation_seconds_batch": 22.285085260984488, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.372091293334961, "genuineness": 4.281409740447998, "identity": 0.4529082179069519, "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_AUTH_reg/3a576102b8dab43090e1abca259c1abcd50c6c0b5192a566d64a65251123adbc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "3acf1950674bac0d1537d1e1517bbf414e1f92cae9b4109e3fc2b02acf3fb0f5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "970e90eb8b52bab9da321ef4e9666b34a7061b9aba132066641a0e62ba24c0ad", "generation_seconds_batch": 27.349145584972575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.268836498260498, "genuineness": 4.15588903427124, "identity": 0.25360727310180664, "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_AUTH_reg/3acf1950674bac0d1537d1e1517bbf414e1f92cae9b4109e3fc2b02acf3fb0f5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "3aef64ab0fc84123b3fb9c236b6b20cacf65a874c5971316fd57f68787262c2c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8aad7ffdd0dbee20af706498fde2c46507354f056d2e62543134a372961e26ae", "generation_seconds_batch": 18.83484937006142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.205295085906982, "genuineness": 3.9550557136535645, "identity": 0.1265369951725006, "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_AUTH_reg/3aef64ab0fc84123b3fb9c236b6b20cacf65a874c5971316fd57f68787262c2c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "3b488279eca1b4170e5b39861498e5a9df89352aaa786cf193baccfb5baac13f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b98f07ebf81e6a37304580e01117e2274b4c35aa7c4e947da97beea4a5aa794", "generation_seconds_batch": 25.589322187006474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.575014352798462, "target_z": 1.312499992828237, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.5393356680870056, "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_AUTH_reg/3b488279eca1b4170e5b39861498e5a9df89352aaa786cf193baccfb5baac13f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "3b938439a11033f1159423b5a80341d6e58ccbd57cde1cad5a20a1ffd0716662", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d351c98f6b813a5c1eb0488b16cb5071363541fb7149121050f3c15edbddf11a", "generation_seconds_batch": 24.645064475014806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.0146708488464355, "genuineness": 4.042920112609863, "identity": 0.7475827932357788, "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_AUTH_reg/3b938439a11033f1159423b5a80341d6e58ccbd57cde1cad5a20a1ffd0716662.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "1285cf95340090368beca819c8dc1233dc0b50e2d5806c7c33bb3bbec5c45b4a", "generation_seconds_batch": 20.008315172046423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.252951145172119, "genuineness": 4.362998008728027, "identity": 0.4842880964279175, "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_AUTH_reg/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "3cc3d50089848e37ed3b10cf3d8a8fd728b8531f42841519bba5d70913a321bb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19b195eb9cfd07825857bdb9d487365c058ff41d568e9b2c8e770ed74b12392a", "generation_seconds_batch": 20.063804444973357, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 6.117925643920898, "genuineness": 4.281409740447998, "identity": 0.8665165901184082, "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_AUTH_reg/3cc3d50089848e37ed3b10cf3d8a8fd728b8531f42841519bba5d70913a321bb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "3d94bd8810ee7b887716213ac07ef05cf59b19d55a2d22e9cb157aa01df02b92", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c6d3abc09d20ce79b886970df5849565864025fe7221aacd8ffbde96a2c0cea", "generation_seconds_batch": 26.108274420024827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.558781862258911, "target_z": 1.2968749587623625, "ce": 6.316492557525635, "genuineness": 4.030367851257324, "identity": 0.3256305158138275, "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_AUTH_reg/3d94bd8810ee7b887716213ac07ef05cf59b19d55a2d22e9cb157aa01df02b92.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "3d9bdf2d2ab59971c9aec53bdc9eee33e9a85b8467ac623fb7391dfe6c89b905", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb44ca76abb71907aee815c28f9732f1740cbf698a159b578f3a6deb3b3ac44c", "generation_seconds_batch": 25.365654174005613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0393433570861816, "target_z": 0.7968750161364668, "ce": 6.300607204437256, "genuineness": 4.18099308013916, "identity": 0.593916654586792, "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_AUTH_reg/3d9bdf2d2ab59971c9aec53bdc9eee33e9a85b8467ac623fb7391dfe6c89b905.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "3da74dbbe2ee61d2f485f82d08efee9ae77833211bb22e6f3892f7e2cdf6e03d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "69c622271f04d671d24a18adc5ec8815e89e5499b8e4a92557d37fc0f47689e8", "generation_seconds_batch": 27.391947184922174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.510084629058838, "target_z": 1.2500000860611564, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.6427558064460754, "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_AUTH_reg/3da74dbbe2ee61d2f485f82d08efee9ae77833211bb22e6f3892f7e2cdf6e03d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "3e2b78d2a8192703bc6c215a72fdc617a6c1394666721682eed7ac74f422bd8f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fcb2fac6607cbac66eb64f434661c64d94dfd4483ef9268963ccada1f51a004", "generation_seconds_batch": 26.27278962091077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.09409761428833, "genuineness": 3.936227321624756, "identity": 0.6074596047401428, "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_AUTH_reg/3e2b78d2a8192703bc6c215a72fdc617a6c1394666721682eed7ac74f422bd8f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "3ec1205fb1a4edf330848e43d532e27c2e477bb2b340b85fb15b418359b3a9f1", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "0745bccadb93f105595804f60a2072f8eef5833ce79de0f1d37115acbca1ba52", "generation_seconds_batch": 19.012173319002613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.045755386352539, "target_z": 1.7656248332565094, "ce": 6.221180438995361, "genuineness": 4.3881025314331055, "identity": 0.5446752309799194, "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_AUTH_reg/3ec1205fb1a4edf330848e43d532e27c2e477bb2b340b85fb15b418359b3a9f1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "3fb096cde575ef86711f3c8c4e4013fc9af518abe7928b3b8ab96d33fe5a2656", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c97cf35e21297c450e0204786696ae1f98e75723671ffdf645dd03b7b656aaad", "generation_seconds_batch": 19.45925772492774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 6.260893821716309, "genuineness": 4.545003414154053, "identity": 0.7586174011230469, "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_AUTH_reg/3fb096cde575ef86711f3c8c4e4013fc9af518abe7928b3b8ab96d33fe5a2656.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "41cdd389c2ec9b0bd438aef8703f71a5b70447988ce492195a24ab11d39f3d4f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc7ded695e3ee132e944f674ce95713de872db4e987ceb46e2b254045a775191", "generation_seconds_batch": 26.72619521606248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.348263263702393, "genuineness": 3.923675537109375, "identity": 0.41041886806488037, "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_AUTH_reg/41cdd389c2ec9b0bd438aef8703f71a5b70447988ce492195a24ab11d39f3d4f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "420b83620f9b171d4459b26492baae3944cf29e630ccae046fc413701dc71e31", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05dcd1cf20e60e81d2205f6da9ac8117b19b14d2c3da91e5bf49fb1e965fb9b7", "generation_seconds_batch": 26.036088299006224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.23706579208374, "genuineness": 3.9927117824554443, "identity": 0.14604324102401733, "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_AUTH_reg/420b83620f9b171d4459b26492baae3944cf29e630ccae046fc413701dc71e31.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": "4232cf39e1f52d8ae994d765053e11b03071b4fb3929bc4185cb24a3676752b9", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49e0863ffc654a87ffa4cae1327c85e5ad74941af0c36dca53a4d6134f0b44a2", "generation_seconds_batch": 19.581194686004892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.41180419921875, "genuineness": 4.030367851257324, "identity": 0.45853137969970703, "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_AUTH_reg/4232cf39e1f52d8ae994d765053e11b03071b4fb3929bc4185cb24a3676752b9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "424ac6c5b1025cba29a9e3c74ae7ad816a5945848e58c34acf4944f58d55db0e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e2c568e491771180e75f5e0c92900d89c453925376886eccc9cb7ca30943ab9", "generation_seconds_batch": 24.56267498107627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.388340950012207, "target_z": 1.1328123305670983, "ce": 6.141753673553467, "genuineness": 4.482243061065674, "identity": 0.8600093126296997, "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_AUTH_reg/424ac6c5b1025cba29a9e3c74ae7ad816a5945848e58c34acf4944f58d55db0e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "428197a3f45093aae4e06deb86c662d003c46b1e23b1006ccde9bb319f48a729", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1db913af7dabb4822d8df0db23f0f4b46fc28750a72c44c3384d0f68963cbc38", "generation_seconds_batch": 20.00900287204422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.316492557525635, "genuineness": 4.494794845581055, "identity": 0.6577020883560181, "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_AUTH_reg/428197a3f45093aae4e06deb86c662d003c46b1e23b1006ccde9bb319f48a729.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "42a10c144754c90545a6a445f9c3be6dc79c05712b95f97341769bc1f73a970a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d5414f4f80793931bdec769620749e272483670f60ced7f03bc4deda2491331e", "generation_seconds_batch": 28.09765619703103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.263189792633057, "target_z": 2.9375000932329196, "ce": 6.1020402908325195, "genuineness": 3.8420867919921875, "identity": 0.6070618033409119, "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_AUTH_reg/42a10c144754c90545a6a445f9c3be6dc79c05712b95f97341769bc1f73a970a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "4320d63cab31067b95c79982ed104c7c6e7eee4b9f19775b54b4944f84a139a5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ac471d7fd567387914f2298a877de667f2525b50d63413d1af7105d0ff78425", "generation_seconds_batch": 25.84204577410128, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7715078592300415, "target_z": 0.5390625277905817, "ce": 6.53094482421875, "genuineness": 4.545003414154053, "identity": 0.38024473190307617, "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_AUTH_reg/4320d63cab31067b95c79982ed104c7c6e7eee4b9f19775b54b4944f84a139a5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "43408654486cdc9fc4a369d3e1ac1d0fadff9a0bf75b313256da298f29ba3870", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c34d966d99a284504738cfc91d14d91d75ac3a8097376f6a4d4998f813099cf9", "generation_seconds_batch": 28.110335026984103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6155953407287598, "target_z": 1.3515623484965058, "ce": 6.022613525390625, "genuineness": 4.262581825256348, "identity": 0.7025187611579895, "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_AUTH_reg/43408654486cdc9fc4a369d3e1ac1d0fadff9a0bf75b313256da298f29ba3870.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "434b37dec0fadb9fed144b5b56c3d0ac24288cd190a1c5666cbca6f7fda0b61d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60a2062a48c7f207c81543d33a21c756ec9fc6ac01251cbdea4cad9047738875", "generation_seconds_batch": 25.83938662696164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.213237762451172, "genuineness": 4.124508857727051, "identity": 0.7075225114822388, "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_AUTH_reg/434b37dec0fadb9fed144b5b56c3d0ac24288cd190a1c5666cbca6f7fda0b61d.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "bc9581537dcce32fab9e4f111a70431446409d9ea41fbf20fb22af63853da69b", "generation_seconds_batch": 18.107821139972657, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.252951145172119, "genuineness": 4.381826400756836, "identity": 0.7868911027908325, "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_AUTH_reg/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "439a629020c0dd20df0e9fd269b907d689820b83bcebb334f28e0df27289dc0e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aef6d9267d87da4811c8870d88f32ff4d44b513c9dd984c0f016eb3828842351", "generation_seconds_batch": 24.335674518020824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.295654773712158, "target_z": 2.9687501613646683, "ce": 6.173524379730225, "genuineness": 4.005263805389404, "identity": 0.6047400236129761, "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_AUTH_reg/439a629020c0dd20df0e9fd269b907d689820b83bcebb334f28e0df27289dc0e.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "b5f485b95951149d1c36218ff3da9afa42b374a27e93aa8624d7f68441ee9f17", "generation_seconds_batch": 18.605552700930275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7129902839660645, "target_z": 1.4453125528917523, "ce": 6.1894097328186035, "genuineness": 4.400654315948486, "identity": 0.5567398071289062, "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_AUTH_reg/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.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": "4438d32e5edcf41c89345e22660d2a62b26cdce060bb21746353e3373d27c34d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1cacbc25b68e72a35130f1350e2564353eb2dffd24338134a7caab8682252ef3", "generation_seconds_batch": 16.732476220000535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.1894097328186035, "genuineness": 4.369274139404297, "identity": 0.786333441734314, "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_AUTH_reg/4438d32e5edcf41c89345e22660d2a62b26cdce060bb21746353e3373d27c34d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "44994a64bce9fa1790cdf6bccc82cbd666114d77ab3869b573d3d8c8f0158fdf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ff609e38fbdf3e0c61585d447a10124ffa150411a9bc377da8e85ba8d909d21", "generation_seconds_batch": 24.814471708028577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.086336612701416, "target_z": 1.8046874184211954, "ce": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.6242591142654419, "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_AUTH_reg/44994a64bce9fa1790cdf6bccc82cbd666114d77ab3869b573d3d8c8f0158fdf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "44a49074e69043ca3ac162fec5a30036cebb8d69434c66079d17bb0a4d1ed332", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2f63d3850525a071b3fddad5f3ac286c0131be6d402ff8751da35b3e7905f195", "generation_seconds_batch": 18.591826800024137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.110685348510742, "target_z": 1.828124969520007, "ce": 6.205295085906982, "genuineness": 4.369274139404297, "identity": 0.7755587100982666, "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_AUTH_reg/44a49074e69043ca3ac162fec5a30036cebb8d69434c66079d17bb0a4d1ed332.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "44a6582ba33f3f0963a1ef39881fbcde672b3fa323c9981dffeecc4a2be7be92", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfe5762992f145e002b70d6673265427ea65c049cb6f2e448ad8a67ad5e79943", "generation_seconds_batch": 18.6703474749811, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 5.95112943649292, "genuineness": 3.8169827461242676, "identity": 0.48925578594207764, "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_AUTH_reg/44a6582ba33f3f0963a1ef39881fbcde672b3fa323c9981dffeecc4a2be7be92.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "44ca21176df458ac99d3853ad1e61226e4e09707c51e0a6f3a35d693bb3adbe4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e648bba80f8fc3cf91ae06eba8ac4157b5c15f8360aa7882e7209ad6f98651ec", "generation_seconds_batch": 24.350154315005057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.173524379730225, "genuineness": 4.444586753845215, "identity": 0.7433739900588989, "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_AUTH_reg/44ca21176df458ac99d3853ad1e61226e4e09707c51e0a6f3a35d693bb3adbe4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "4538b66dd5bafb28b2004b0e27864fcf2697d2952f3645089b53a521dbbb4cae", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8914d8eacde406c8f42ccb462ee2a8c27997e6b86e8e8bb03df9722a338b9be1", "generation_seconds_batch": 25.940952660981566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.316492557525635, "genuineness": 4.061748504638672, "identity": 0.1698608547449112, "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_AUTH_reg/4538b66dd5bafb28b2004b0e27864fcf2697d2952f3645089b53a521dbbb4cae.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "455adc470d0f0e936c42375bca21ee91d97e85b8413ab0e83395cd52473c2a7e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cd010954488ac00d666687935f4f2f2c259cbcea1760ffbf323611a98713b37", "generation_seconds_batch": 25.90016617393121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.372091293334961, "genuineness": 4.250029563903809, "identity": 0.6737815737724304, "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_AUTH_reg/455adc470d0f0e936c42375bca21ee91d97e85b8413ab0e83395cd52473c2a7e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "45a6326272fd3d8385c9793185e4559ed9dcf127cc2f6dbcb9b96f66abdc546b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84be7c91289264ced74393f31b55eef9b34bc72ca81b68c5f211a092042ad58e", "generation_seconds_batch": 27.349145584972575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2665977478027344, "target_z": 1.0156250340658743, "ce": 6.332377910614014, "genuineness": 4.394378662109375, "identity": 0.46719449758529663, "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_AUTH_reg/45a6326272fd3d8385c9793185e4559ed9dcf127cc2f6dbcb9b96f66abdc546b.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "8d810f67b2d52ec9511b176398f73b53e0fefa5ea931b70f09581e56501fac32", "generation_seconds_batch": 20.484106425079517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.340320587158203, "genuineness": 4.042920112609863, "identity": 0.5376344919204712, "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_AUTH_reg/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.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": "46617e530bb740ab019839359bf49a8ce5886ab08cfa962589ef743f1ea1a354", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "45e7852a50a9d36ca2ac3b664f200702354575a1db93d2e7935668eec260cfb2", "generation_seconds_batch": 22.408172048046254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.284721851348877, "genuineness": 4.042920112609863, "identity": 0.311053603887558, "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_AUTH_reg/46617e530bb740ab019839359bf49a8ce5886ab08cfa962589ef743f1ea1a354.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "46acad19b99b33210bdc13cc2675270d1cb4e106475de54ab95f35b9f020e4df", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ebf3228d2b304aa64f85f88c6336388202c5fe2512ed08b9c44a3fc010a83b2", "generation_seconds_batch": 25.666995398001745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.316492557525635, "genuineness": 4.306513786315918, "identity": -0.05637899413704872, "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_AUTH_reg/46acad19b99b33210bdc13cc2675270d1cb4e106475de54ab95f35b9f020e4df.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "472d61019c1d2111d02a4e0a45e51a06fc71509c34ba1ee2539ee7d1e8894258", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6204c5dba4ecaa5669c41030feb9ebaab2315f2a0b721640dbcaa43e7ea97dd", "generation_seconds_batch": 24.647520172060467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.340320587158203, "genuineness": 4.080576419830322, "identity": 0.492729127407074, "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_AUTH_reg/472d61019c1d2111d02a4e0a45e51a06fc71509c34ba1ee2539ee7d1e8894258.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "47330525f8466e27e40b6a3cdb43d4bb070839002d0315f110bfe973a67678b1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68cc2a6e0dd414a70f226a08dff8cb93138466158583d37fd46ffdb204a21ba3", "generation_seconds_batch": 23.91770193609409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.459460258483887, "genuineness": 4.381826400756836, "identity": 0.5149840116500854, "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_AUTH_reg/47330525f8466e27e40b6a3cdb43d4bb070839002d0315f110bfe973a67678b1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "475e5d82521b935fb90081ffafff797a440315fe8d91456e976e09a320b79c2b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32165e1c2877e0a2eb1008d72e0cfbdf29e3842dc182518f77175ab4ef58262b", "generation_seconds_batch": 24.732398508000188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.23706579208374, "genuineness": 4.306513786315918, "identity": 0.6705684065818787, "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_AUTH_reg/475e5d82521b935fb90081ffafff797a440315fe8d91456e976e09a320b79c2b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "47788e19b00851a118a0b58c0318a058e04a58132018068ac460683c3f5fd9e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "626ebbdcb2a3e098829408037b9b6eeba4e046a91404bc0f413a1f4217e524b9", "generation_seconds_batch": 26.923399708000943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.23706579208374, "genuineness": 4.319066047668457, "identity": 0.45853352546691895, "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_AUTH_reg/47788e19b00851a118a0b58c0318a058e04a58132018068ac460683c3f5fd9e5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "47883c0864aa20e636df1495b2d3a0ea0a8abf005df5b81bf69831acf111a344", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8ebef98a8ce852c3ded786d2cffb9afcfc81f4ef52950760727a51d7b140d6a", "generation_seconds_batch": 21.852288442081772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6967577934265137, "target_z": 1.429687518825878, "ce": 5.974957466125488, "genuineness": 4.319066047668457, "identity": 0.528910756111145, "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_AUTH_reg/47883c0864aa20e636df1495b2d3a0ea0a8abf005df5b81bf69831acf111a344.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "f13631d0ef5fff213e10ffa3fb3b0f3a10e2fc259dfd5ceb232fb917b76bda9f", "generation_seconds_batch": 18.483687986037694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.173524379730225, "genuineness": 4.3127899169921875, "identity": 0.7497626543045044, "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_AUTH_reg/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "47fec22e1b54243b037aa646e6598199c6143224c64beb75c059c6d4dcaffa06", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68ecd4bd7b0ce743c11ff23e04db0fbe7f551d7a22ab6950b8f3aa0db0477564", "generation_seconds_batch": 25.743776886025444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 6.109982967376709, "genuineness": 3.9927117824554443, "identity": 0.10297759622335434, "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_AUTH_reg/47fec22e1b54243b037aa646e6598199c6143224c64beb75c059c6d4dcaffa06.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "48635f21906328f852eba94b096db2837936f57240c842b1184526d1d9a7004b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c578bdbfdf47e7be0e93aac8e8d84f67fff9a45871d0278f91abf15cac902f87", "generation_seconds_batch": 19.48725551192183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.149696350097656, "genuineness": 3.9676074981689453, "identity": 0.31083494424819946, "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_AUTH_reg/48635f21906328f852eba94b096db2837936f57240c842b1184526d1d9a7004b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "489a782bc099a5512dcb80fb2f02b1faabfc0d83ab21efdfb79793032b6ad0c5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cac65f585cc975b7c9e54354b55fbd32a09dfbb54d3c912f5486af1e56b3227", "generation_seconds_batch": 26.578145529027097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0684003829956055, "target_z": 2.7500001434352606, "ce": 6.181467056274414, "genuineness": 4.20609712600708, "identity": 0.6717519760131836, "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_AUTH_reg/489a782bc099a5512dcb80fb2f02b1faabfc0d83ab21efdfb79793032b6ad0c5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "48d04e7494546b66c6a4106363363d8406ad64199a40e5f1a14803bbb82b6156", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5877f75a1c0fc18d4368eeaf06a0524c8aa498b7dded0ed9756a328c6db6d7a3", "generation_seconds_batch": 18.108147385064512, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.316492557525635, "genuineness": 4.3881025314331055, "identity": 0.5142688751220703, "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_AUTH_reg/48d04e7494546b66c6a4106363363d8406ad64199a40e5f1a14803bbb82b6156.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "4c4d56e8a48394a3d2f3cf0761bd0b7d017fc745cfcb7da3eba0600fdd432429", "generation_seconds_batch": 16.36132884491235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.24500846862793, "genuineness": 4.030367851257324, "identity": 0.6112837791442871, "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_AUTH_reg/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "4af5c11ef750e367c1014b7feba1f19eeaf81123e4655f9978681f8a4dd88755", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a18dc94ea145f8652597ca83077e2ea8c578363e898811e6f38c380caaa957bc", "generation_seconds_batch": 25.13642307196278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.021406888961792, "target_z": 1.742187511654115, "ce": 6.356205940246582, "genuineness": 4.287685871124268, "identity": 0.3327710032463074, "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_AUTH_reg/4af5c11ef750e367c1014b7feba1f19eeaf81123e4655f9978681f8a4dd88755.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "4b257af0461ad9bad4a366c00f05f376405ff7b163756bde2d0244f77bc1f87a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8f05982ceb7bca29c1ff7b7523b06c2d3eaee6677bd9b44994d9df7e4e8cdac3", "generation_seconds_batch": 17.954380632028915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4857358932495117, "target_z": 1.2265625349623448, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.6975181698799133, "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_AUTH_reg/4b257af0461ad9bad4a366c00f05f376405ff7b163756bde2d0244f77bc1f87a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "4b8a6dfffe9230297813422c963756def9250624a9135436c497916da57e308e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33cae0e33ba139c7254365e83b20795444ad0edcfd6668e31e8a7810d96e5854", "generation_seconds_batch": 22.802018410991877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8671984672546387, "target_z": 1.5937499175247252, "ce": 6.173524379730225, "genuineness": 4.256305694580078, "identity": 0.6542379260063171, "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_AUTH_reg/4b8a6dfffe9230297813422c963756def9250624a9135436c497916da57e308e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "4c2b3dc47b5707dc61212d627cefa920cefcda5cbcca76732a339c0b3fdaf3ef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "29dd1214958267ae36d01d22dccd9d732c829b86db289661dd0403acd1bb3262", "generation_seconds_batch": 25.585969454958104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9662972688674927, "target_z": 0.7265624775882406, "ce": 6.4912309646606445, "genuineness": 4.545003414154053, "identity": 0.37937718629837036, "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_AUTH_reg/4c2b3dc47b5707dc61212d627cefa920cefcda5cbcca76732a339c0b3fdaf3ef.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": "4c57c5607abac0db1ef761b6b6a3910ea5c58e97cb0c6b6ff3900438984f54b8", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "986ab60a16497b7369ed80439b4ad5d3ac76077ddfd9c5a5cb9d240d7cc5982d", "generation_seconds_batch": 26.15458785602823, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.117925643920898, "genuineness": 4.005263805389404, "identity": 0.3833933472633362, "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_AUTH_reg/4c57c5607abac0db1ef761b6b6a3910ea5c58e97cb0c6b6ff3900438984f54b8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "4c6f9c8f2d2d62ef3ec2b8442f3c2d1e232f307ad9bf0dcec622dab99f90f4bb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9623e90b6b79c02e3640ad7dfa4e60839562e6762570c26b7643f191979d2485", "generation_seconds_batch": 25.30643321503885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6480603218078613, "target_z": 1.3828124166282547, "ce": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.6021753549575806, "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_AUTH_reg/4c6f9c8f2d2d62ef3ec2b8442f3c2d1e232f307ad9bf0dcec622dab99f90f4bb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "4ca122f9b8bcc0f9c15316bfd9446026c8f39cc76ea5fddc6ca27eed1537183f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "701cc79edf6a7b5009a77e3abac1c96251ca87e4bcd69f38fa975304974e94cb", "generation_seconds_batch": 27.391947184922174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.623711585998535, "target_z": 1.359374865529443, "ce": 5.911416053771973, "genuineness": 4.306513786315918, "identity": 0.6824727058410645, "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_AUTH_reg/4ca122f9b8bcc0f9c15316bfd9446026c8f39cc76ea5fddc6ca27eed1537183f.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": "4d1e2575dfd5ab87f3346e54e2a8ef28b57b52b767acf37a2a877be73380fe64", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ae2e338c82891b8c46b211d73b5dcf786da182f28463776b7104e0f4eee31f12", "generation_seconds_batch": 16.670227741939016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.5296790599822998, "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": 0.120000000000001, "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_AUTH_reg/4d1e2575dfd5ab87f3346e54e2a8ef28b57b52b767acf37a2a877be73380fe64.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": "4d282535e527c041ab9af66b66e3a3d61e5852dffaf63861a463b3b94da235a0", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aeb25420b3778179de4be303ffd36348c918a48290dc73c3dd0f0c4fde5e4d35", "generation_seconds_batch": 16.499336790060624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0684003829956055, "target_z": 2.7500001434352606, "ce": 6.24500846862793, "genuineness": 4.068024158477783, "identity": 0.6169130802154541, "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_AUTH_reg/4d282535e527c041ab9af66b66e3a3d61e5852dffaf63861a463b3b94da235a0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "4db2cc68ce36b68582ba274e694c5929f67ff1d051f7dbe1df6155092974e45b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df6d22507a95f278e5ba86517a40195a529db9d18123fc78b46d67c830e657b2", "generation_seconds_batch": 28.0878754099831, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0944528579711914, "target_z": 1.8124999354541327, "ce": 6.078212261199951, "genuineness": 3.848362922668457, "identity": 0.7299377918243408, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient à ce qu'elle s'y arrua. J'ai trouvé la petite boîte à mon tiernimet.", "wer": 0.6842105263157895, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/4db2cc68ce36b68582ba274e694c5929f67ff1d051f7dbe1df6155092974e45b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "4dc0045f6e89c3ef1ac7ef92c5b327a8d5e0010d575a6c670fc33976fb7a4e29", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "866016b2e25596efedb63a1f526463b54b7e99353eef1380b55a3ab0402faf67", "generation_seconds_batch": 27.719575587077998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.38003396987915, "genuineness": 4.011539936065674, "identity": 0.6740303635597229, "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_AUTH_reg/4dc0045f6e89c3ef1ac7ef92c5b327a8d5e0010d575a6c670fc33976fb7a4e29.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "4e7d8a2d1fe6594b5ec4c4d2075cf92f98c3f56c4199cdacead3625a6f542bc1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26bd5a98d5af492b27917b4e28ad04060b1cab7349213c28f0b563f883cd14b0", "generation_seconds_batch": 26.081687061931007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.664292812347412, "target_z": 1.398437450694129, "ce": 5.847874641418457, "genuineness": 4.394378662109375, "identity": 0.8478778600692749, "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_AUTH_reg/4e7d8a2d1fe6594b5ec4c4d2075cf92f98c3f56c4199cdacead3625a6f542bc1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "4e990926d93921e0eb113914eea22d39d48f5b22b57114a3ef1e3945ccc80e49", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ea18272d154eda952349613fa2b262cb0ae52582c6efe91028f25a5333488a1", "generation_seconds_batch": 25.537519666017033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.038498878479004, "genuineness": 4.005263805389404, "identity": 0.6208544969558716, "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_AUTH_reg/4e990926d93921e0eb113914eea22d39d48f5b22b57114a3ef1e3945ccc80e49.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "4f2ae35c54034ff3399f35800e2d77f8a0c32695e58e336fe4eea6c912fe721f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "589b79781c3a73ef6612e5845912a4644304cef1b70723787c3b8da9d75f68c7", "generation_seconds_batch": 25.16831897792872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7129902839660645, "target_z": 1.4453125528917523, "ce": 5.95112943649292, "genuineness": 4.281409740447998, "identity": 0.7156230211257935, "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_AUTH_reg/4f2ae35c54034ff3399f35800e2d77f8a0c32695e58e336fe4eea6c912fe721f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "4f9ee6fb01bf59d886167639178b5aa2e90730491adc8fb6b87db0245a65c5e3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87d957d5ef225918907becd3655d5d7f3913e40fbf5aa593bb6db3f960649fa6", "generation_seconds_batch": 18.687926208018325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.221180438995361, "genuineness": 4.419482707977295, "identity": 0.64288330078125, "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_AUTH_reg/4f9ee6fb01bf59d886167639178b5aa2e90730491adc8fb6b87db0245a65c5e3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "50cb1289378aabdc8c68f6df0b97936b8cf56246dbe5e693aba2a633805947d9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "264e427e0e9711a74d02cc049fbabe8b1b4ba869aa01006e9def34809e210202", "generation_seconds_batch": 26.34820273704827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.300607204437256, "genuineness": 4.293961524963379, "identity": 0.2981864809989929, "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_AUTH_reg/50cb1289378aabdc8c68f6df0b97936b8cf56246dbe5e693aba2a633805947d9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "50d95d2fbeb1ec267d06ffec658472ba23dacdfdc494552707f1a240e10f0dfa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a6d59029fedba0b60e46406452ea2a279bfc48d248d3ab8508225066cea71dfd", "generation_seconds_batch": 19.2274131529266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8770188093185425, "target_z": 0.6406250197223483, "ce": 6.340320587158203, "genuineness": 4.482243061065674, "identity": 0.4299485385417938, "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_AUTH_reg/50d95d2fbeb1ec267d06ffec658472ba23dacdfdc494552707f1a240e10f0dfa.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": "50dfe61dfb05a5bc76d34bb494b75f450f3be622b1fe3540f94aaaf2e52ac373", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "4098fafa2c5ca3bbf07cbaf3f455c37d40dc9c0d287f25161f66adca62beba15", "generation_seconds_batch": 16.670227741939016, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149562358856201, "target_z": 2.8281248547717985, "ce": 6.2767791748046875, "genuineness": 4.080576419830322, "identity": 0.6601788997650146, "hypothesis": "Is schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.043478260869565216, "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_AUTH_reg/50dfe61dfb05a5bc76d34bb494b75f450f3be622b1fe3540f94aaaf2e52ac373.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "5106457f8bad3ec20eaedb03b72959e7c68406cc6d7aea5c10cfe428739fe272", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4384cfc4ee54bb596849df6a951a95577d697e9341aceefc8b18f11beb4c72fa", "generation_seconds_batch": 21.399306531995535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.100865364074707, "target_z": 2.7812502115670097, "ce": 6.197352409362793, "genuineness": 4.306513786315918, "identity": 0.6990699768066406, "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_AUTH_reg/5106457f8bad3ec20eaedb03b72959e7c68406cc6d7aea5c10cfe428739fe272.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "513783edf7a8384ff67d0c7e75a4d9acb34972c5d42185d821bf5e06ff8f1f91", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acf8e7fe543922057853c5471ae790bc124f72af96dd391ca51b211b77924c49", "generation_seconds_batch": 25.142816429026425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3964571952819824, "target_z": 1.1406248476000356, "ce": 6.022613525390625, "genuineness": 4.3504462242126465, "identity": 0.5014711618423462, "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_AUTH_reg/513783edf7a8384ff67d0c7e75a4d9acb34972c5d42185d821bf5e06ff8f1f91.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "515463d4d8def9339397cb04022e3bbb512782716de59dae9f8a4f4310b49ef4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b2d7ef7bccb4c0c1703e08ee3238d672ca1643f20aa321130f329a4fda1b24d", "generation_seconds_batch": 25.477295704069547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.260893821716309, "genuineness": 4.444586753845215, "identity": 0.7111495733261108, "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_AUTH_reg/515463d4d8def9339397cb04022e3bbb512782716de59dae9f8a4f4310b49ef4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "5161c2a8f06f8ec14ad0f2ea76c3c57193b6db2533c88d7347a9e36c19d778f7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b13f035e332626db04f20d9891f9a390bca10af24e648daec6cbf5e649067d7", "generation_seconds_batch": 26.69096199190244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.332377910614014, "genuineness": 4.068024158477783, "identity": 0.25767385959625244, "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_AUTH_reg/5161c2a8f06f8ec14ad0f2ea76c3c57193b6db2533c88d7347a9e36c19d778f7.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": "5162bf6beae01882438dcb96492f0a1259ef28fbb9d45d73560b68868dbff0d2", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea063d9a8c023756649eb539d4e66ca07822bd104325c926d00a9e1be9fb4053", "generation_seconds_batch": 22.37587234005332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.38797664642334, "genuineness": 3.9676074981689453, "identity": 0.33407294750213623, "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_AUTH_reg/5162bf6beae01882438dcb96492f0a1259ef28fbb9d45d73560b68868dbff0d2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "51d7e7b5e011cd16c31b1d2456ce488573e9eb0d6b48c3aab4fbe9671dc9e004", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03fbdc116cfe5ad87d02fc5336c022a5ee9b6cacbb9ccb42296d1ddc665d4c9f", "generation_seconds_batch": 19.730083729955368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932128429412842, "target_z": 1.6562500537882228, "ce": 6.300607204437256, "genuineness": 4.111956596374512, "identity": 0.31058764457702637, "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_AUTH_reg/51d7e7b5e011cd16c31b1d2456ce488573e9eb0d6b48c3aab4fbe9671dc9e004.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "5221b27a78a8eb909959a7d24d165efea7a192d0e8e7d26cc458b27363d95de2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74f706c265a115fc540242a83b28cda726eb53c433d91087e31fd4cbfdca147a", "generation_seconds_batch": 18.99400839698501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.875314712524414, "target_z": 1.6015624345576622, "ce": 6.1894097328186035, "genuineness": 4.557555198669434, "identity": 0.7244987487792969, "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_AUTH_reg/5221b27a78a8eb909959a7d24d165efea7a192d0e8e7d26cc458b27363d95de2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "524227aa3ef25afce8cacb7010d923ac6caaa86829edb0d38e4a1c8c306d74ff", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f82235f362818c09f71aece8a47ce935b2c8e2e0652451b7fbc24a34b671ee6a", "generation_seconds_batch": 20.03564080502838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402869701385498, "target_z": 2.1093751237129124, "ce": 6.006728172302246, "genuineness": 3.961331605911255, "identity": 0.7792067527770996, "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_AUTH_reg/524227aa3ef25afce8cacb7010d923ac6caaa86829edb0d38e4a1c8c306d74ff.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "5361908e1b9cd9cdafb3f08a80d7af73bb5ca34736feb38644204cb07cc5103a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "541d427f913a9a242adbc99fd5549f3ba8af69e244a3974b6987f225468c1b1b", "generation_seconds_batch": 23.81994315504562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.230724811553955, "target_z": 2.9062500251011705, "ce": 5.990842819213867, "genuineness": 3.9550557136535645, "identity": 0.5906318426132202, "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_AUTH_reg/5361908e1b9cd9cdafb3f08a80d7af73bb5ca34736feb38644204cb07cc5103a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "5391a91b2971fcde89089db30848c6c77090f0246177c8beb9cb9912419119ee", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1f915e3e63aaedfb5eb67aedfb5f81a1ed17c85bb0c761c0cbfea98a35e4cc5", "generation_seconds_batch": 18.845960006932728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.38797664642334, "genuineness": 4.381826400756836, "identity": 0.6392309665679932, "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_AUTH_reg/5391a91b2971fcde89089db30848c6c77090f0246177c8beb9cb9912419119ee.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "540f72fffb43cbcef87aca580366d8f59518c2ca733d6959fb027cb90fb2e832", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3f92c75ff41e45010f29ae347e8c75939d97a84725f3f89adb8d94d849711b5", "generation_seconds_batch": 26.223303742939606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.263189792633057, "target_z": 2.9375000932329196, "ce": 6.109982967376709, "genuineness": 4.168440818786621, "identity": 0.6784629225730896, "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_AUTH_reg/540f72fffb43cbcef87aca580366d8f59518c2ca733d6959fb027cb90fb2e832.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "5430f7fc929f95a59cf520d1804e9f9798d315de3c8c07b5eb16fad299e795db", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dea422995bf2fd3652d382408b6ba7e8d58fb17f974a8f810990aad9718bb7f", "generation_seconds_batch": 24.923642256995663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.006728172302246, "genuineness": 4.068024158477783, "identity": 0.7138928174972534, "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_AUTH_reg/5430f7fc929f95a59cf520d1804e9f9798d315de3c8c07b5eb16fad299e795db.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "544dd9b6a3d6ea8c337cccf5e92da60a74ffd72a4b0e315fbe87f1920d21a4c3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "80e2a0d0013424ede5646ae62344daf324f14de84129e8acc2d8c0c6c0dc2bd5", "generation_seconds_batch": 26.08336986694485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.786036252975464, "target_z": 1.5156249766917702, "ce": 6.300607204437256, "genuineness": 4.582659721374512, "identity": 0.15283632278442383, "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_AUTH_reg/544dd9b6a3d6ea8c337cccf5e92da60a74ffd72a4b0e315fbe87f1920d21a4c3.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "06fe3296b462695303de3f14d988cb9dde46a39648f1534f1766658b34a8b1fe", "generation_seconds_batch": 20.484106425079517, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.292664527893066, "genuineness": 3.835810661315918, "identity": 0.48480284214019775, "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_AUTH_reg/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "5467902ed8808fa044e9e64fb576762252056471c6833ba56881c86333882cd1", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "f4af14650a7c22395951817ca4e26a02e0c84df41be2aa07b7f77a44c4456e2f", "generation_seconds_batch": 20.067598817055114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1610867977142334, "target_z": 0.9140625421341079, "ce": 6.0305562019348145, "genuineness": 4.469690799713135, "identity": 0.6476276516914368, "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_AUTH_reg/5467902ed8808fa044e9e64fb576762252056471c6833ba56881c86333882cd1.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": "54aa20ad6c67d12f766d24dd46778ca6d488ba19425a3c8246243a481a545503", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a06b250788fa9969dd3e6b554916bd55e42208bd8fe93e02ea6c273a7569832d", "generation_seconds_batch": 16.516120343003422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.133810997009277, "genuineness": 4.2751336097717285, "identity": 0.6848127841949463, "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_AUTH_reg/54aa20ad6c67d12f766d24dd46778ca6d488ba19425a3c8246243a481a545503.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "54ac0f1739f3b828bcb38005a625c018b2f9df6730cd72bdfae6451edb9d833d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "304adbd82fb063c6aaee919d8150bae1a891b298d77392a7b08b02aecc0e808b", "generation_seconds_batch": 18.633984974003397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.157639026641846, "genuineness": 4.118232727050781, "identity": 0.8006507158279419, "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_AUTH_reg/54ac0f1739f3b828bcb38005a625c018b2f9df6730cd72bdfae6451edb9d833d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "5563469b66a6580d0d6e707eba29a096e728b91d84c4da84a6dfea5bb3e0f331", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fb66d6f131efc44c856af4bd18e031b31539f0acedc0f73ed2c86295cfccde8", "generation_seconds_batch": 25.00538860098459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.761687755584717, "target_z": 1.4921876550893756, "ce": 6.1020402908325195, "genuineness": 4.369274139404297, "identity": 0.6156250238418579, "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_AUTH_reg/5563469b66a6580d0d6e707eba29a096e728b91d84c4da84a6dfea5bb3e0f331.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "5582d47b3bda921e80b2dc1755f62754f2e523335b9cef4f33666f43c9d93da4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b14a7e53e9ab00683c09c1a45b0f20af3d877564a7311cec97b355331bca3f63", "generation_seconds_batch": 25.255860126926564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 5.847874641418457, "genuineness": 4.15588903427124, "identity": 0.7713836431503296, "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_AUTH_reg/5582d47b3bda921e80b2dc1755f62754f2e523335b9cef4f33666f43c9d93da4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "55e30013d258dec1fe6fd18706306b83a90b673c260fe476c7ec77f53752458b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d69168a5d539b114de2d23055eb48f28fc35eb85eb3d8ef2f5026c12ef8afcb6", "generation_seconds_batch": 26.779353736899793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.063692092895508, "target_z": 0.8203125672352785, "ce": 6.24500846862793, "genuineness": 4.193545341491699, "identity": 0.674812912940979, "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_AUTH_reg/55e30013d258dec1fe6fd18706306b83a90b673c260fe476c7ec77f53752458b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "55e3ee4d9b5597c98a2db9a1b9e13ddbb36adcdcf8f6fff47c5c050e2a35d2c5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1eeb0af90406a52d76e0132311fbba60d94e31399baa6199bc1c3e61c94997ea", "generation_seconds_batch": 24.26740481692832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.901367425918579, "target_z": 0.6640624560729514, "ce": 6.435632705688477, "genuineness": 4.570107460021973, "identity": 0.3958757519721985, "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_AUTH_reg/55e3ee4d9b5597c98a2db9a1b9e13ddbb36adcdcf8f6fff47c5c050e2a35d2c5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "55e68baab365bbf1712f51ca6f85427dabb50e2b93bebbfbd64f1f7ca4aba8b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6082353fceb2377bb0bec5755144120afb05be014655b4d040e4f2f28a8357ea", "generation_seconds_batch": 25.46088673803024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 6.054384231567383, "genuineness": 4.13078498840332, "identity": 0.703902542591095, "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_AUTH_reg/55e68baab365bbf1712f51ca6f85427dabb50e2b93bebbfbd64f1f7ca4aba8b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "562c482522c37c2a4e4843d7341f279e722f6f1a2e3c053bb484b5e32ad94d00", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d24e3c7b89cc8f319d2fa61bc1f416e575f626ddedea17782c9b77e9d506210", "generation_seconds_batch": 26.352658731979318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.2767791748046875, "genuineness": 4.582659721374512, "identity": 0.7118287682533264, "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_AUTH_reg/562c482522c37c2a4e4843d7341f279e722f6f1a2e3c053bb484b5e32ad94d00.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "567bcd8d9b334e5ba8975ec4a4a94d558b1c604c58b270653e339f50bf547584", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2439136045d3feea5d73d230be06043235ba1087605669cdd7a5c14de0483f8", "generation_seconds_batch": 19.540063317050226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 5.974957466125488, "genuineness": 4.13078498840332, "identity": 0.45840224623680115, "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_AUTH_reg/567bcd8d9b334e5ba8975ec4a4a94d558b1c604c58b270653e339f50bf547584.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "56b9d086923b6926fc85afd84c1d504b296b6b634fd40a20f886e90008d8da1f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8686b78f12164ea2d925ffdf7e40278feca4bcdfc8e117b57d2810542a81d249", "generation_seconds_batch": 27.72684156289324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9223079681396484, "target_z": 2.609374836842391, "ce": 6.117925643920898, "genuineness": 4.168440818786621, "identity": 0.6371551156044006, "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_AUTH_reg/56b9d086923b6926fc85afd84c1d504b296b6b634fd40a20f886e90008d8da1f.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": "56ef8501ec8455d3a81160fd83e54d1558ef87ffd6817a008a8b12bf2c5f97c9", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a20c62219f30bc0cfcc7da7ec414a0c2e4a331c06256bc91ef45a1ef08add07", "generation_seconds_batch": 25.952241010032594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9223079681396484, "target_z": 2.609374836842391, "ce": 6.292664527893066, "genuineness": 3.9927117824554443, "identity": 0.6052429676055908, "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_AUTH_reg/56ef8501ec8455d3a81160fd83e54d1558ef87ffd6817a008a8b12bf2c5f97c9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "576a1728624da3fefbabda35e4ca450837754a089fd731c4039eefbf49dff7f1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d522e596eb4fde6d9e7c9a90cefe75e172703f56febfa04e48f8474b53a9346", "generation_seconds_batch": 21.852288442081772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2405447959899902, "target_z": 1.9531247830541683, "ce": 5.982900142669678, "genuineness": 3.8169827461242676, "identity": 0.7035792469978333, "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_AUTH_reg/576a1728624da3fefbabda35e4ca450837754a089fd731c4039eefbf49dff7f1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "57740d7bb45862049ac01ebbcfa7cc8c09e3457d52e803733886710c6aa32cdf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3925a7adc5c0fc61e443c891132164429bd8e9a133c7ea88a31376cbdbc21923", "generation_seconds_batch": 20.834412113996223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9094836711883545, "target_z": 0.6718749731058886, "ce": 6.165581703186035, "genuineness": 4.68307638168335, "identity": 0.6780110597610474, "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_AUTH_reg/57740d7bb45862049ac01ebbcfa7cc8c09e3457d52e803733886710c6aa32cdf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "57c4a6204ff05499190ee03750b7e7101a9555156a8110cf1dda5ccc25c5fb03", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d9d34a8df1dccbdc0c2bd0cef1723d0597c0b8532787d3e7d7b27a40b7090df1", "generation_seconds_batch": 26.794133082963526, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.086154937744141, "genuineness": 4.362998008728027, "identity": 0.6056786775588989, "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_AUTH_reg/57c4a6204ff05499190ee03750b7e7101a9555156a8110cf1dda5ccc25c5fb03.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "57ce6608b01daa3d45a3a3e6ed8e360df57407bd98861e3ec6b8efc94b92e2a2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b067d1f5b8ddd7712c0b59e1ebebf3308254f1ba314a251e827a65574df76d7", "generation_seconds_batch": 25.959222548059188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.054384231567383, "genuineness": 4.118232727050781, "identity": 0.06909219175577164, "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_AUTH_reg/57ce6608b01daa3d45a3a3e6ed8e360df57407bd98861e3ec6b8efc94b92e2a2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "585cb78f97a70faae061f6bff8e8817faaddcb2a2e8272c5215b21e6843ca07e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1d21e3a9ab9d1bd265e5f2b30650d34de61c84c0a8bf36fbf7976d225eb1076", "generation_seconds_batch": 25.603032885934226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 5.4825119972229, "genuineness": 3.7165660858154297, "identity": 0.3485479950904846, "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_AUTH_reg/585cb78f97a70faae061f6bff8e8817faaddcb2a2e8272c5215b21e6843ca07e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "587b6678753f7cb6e5bedbed28ede722aad9336905a1218b6c3aa35722db3173", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06cefbfdf0927d46976b73702d9a56e029bf3153b9a2a917b5ef2463ef2a3446", "generation_seconds_batch": 25.763939644908533, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.2767791748046875, "genuineness": 4.030367851257324, "identity": 0.08444350957870483, "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_AUTH_reg/587b6678753f7cb6e5bedbed28ede722aad9336905a1218b6c3aa35722db3173.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "59387b58745a7927d8b1d2db8a92850ccf8b4249f087a921a01831f7e0c76c96", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e636c5127d1cfe67249a2aec3d4a306d6570d2d264c1d3ad0482823e14e4205e", "generation_seconds_batch": 19.359179602004588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0538716316223145, "target_z": 1.7734373502894467, "ce": 6.125868320465088, "genuineness": 4.262581825256348, "identity": 0.5010774731636047, "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_AUTH_reg/59387b58745a7927d8b1d2db8a92850ccf8b4249f087a921a01831f7e0c76c96.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "59ba1072df620c5607954061cdd8974cffb6e8f218851a8315ef51f75ebcc39a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f993bb200a69d3b6e52a67304681c05db841dbfd0937694acfd37cbc2f5ede01", "generation_seconds_batch": 26.15257835597731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8022687435150146, "target_z": 1.5312500107576446, "ce": 6.38797664642334, "genuineness": 4.582659721374512, "identity": 0.5921325087547302, "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_AUTH_reg/59ba1072df620c5607954061cdd8974cffb6e8f218851a8315ef51f75ebcc39a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "59cede3012a2ee53551e1e97f73d09bf09405e109862a34d26e03252b6d76171", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe1e1f0f44c9379c3de2b190d4c3986d5172e49d5099331335d26b02015e2fe7", "generation_seconds_batch": 19.467890167958103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.639944076538086, "target_z": 1.3749998995953174, "ce": 6.070269584655762, "genuineness": 4.444586753845215, "identity": 0.5934826731681824, "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_AUTH_reg/59cede3012a2ee53551e1e97f73d09bf09405e109862a34d26e03252b6d76171.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "59f7885b57591572f40b28834eeef97723e9ee15d521378e4d0087df12f0de7a", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "5ef6f893e71d446435c065ad0f53de84b2bf2b98084c8ffd43ba7a045cddfcf2", "generation_seconds_batch": 20.150925190071575, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.109982967376709, "genuineness": 4.394378662109375, "identity": 0.7989578247070312, "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_AUTH_reg/59f7885b57591572f40b28834eeef97723e9ee15d521378e4d0087df12f0de7a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "5a86fd37b555313501f96b76dd5be93acd83ad3be1371a88de6a50d085079b5c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "37fa8e93d9d29465d9b7582c0da02de4caf91f71eaaf13fd9ad7168d65c88d4f", "generation_seconds_batch": 24.241931298049167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8590822219848633, "target_z": 1.5859374004917879, "ce": 6.221180438995361, "genuineness": 4.250029563903809, "identity": 0.4103783369064331, "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_AUTH_reg/5a86fd37b555313501f96b76dd5be93acd83ad3be1371a88de6a50d085079b5c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "5ae2b3a0c1d23f06638721a92030f473fb1e9f1977bfa620382ab8bbc65054aa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12abc83a25eed86f589e36367e10d7c4dd8c79c52603f48814bd92e4da0be19e", "generation_seconds_batch": 24.044044145033695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.260893821716309, "genuineness": 3.860914945602417, "identity": 0.37562236189842224, "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_AUTH_reg/5ae2b3a0c1d23f06638721a92030f473fb1e9f1977bfa620382ab8bbc65054aa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "5b68cc4d7da966b824ac37f08585772485143b1300d3b09b9fa08141a97ccf9f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b031b44067c84698470607ca68cc1d5a3a49159476ec8552922ff25d9c641371", "generation_seconds_batch": 24.690683279070072, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 5.788304805755615, "genuineness": 3.8671910762786865, "identity": 0.5759591460227966, "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_AUTH_reg/5b68cc4d7da966b824ac37f08585772485143b1300d3b09b9fa08141a97ccf9f.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": "5b6db489d2019fc9bb480778b514ef7fa5cdff6318deb91886eb438fae5eb8f5", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f72b24e5664dcdde40d5c75b4ee46b31da3ce749ce3e2ff029c10e66bafdc8bd", "generation_seconds_batch": 22.146845887997188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.630124092102051, "target_z": 2.32812514164232, "ce": 6.419747352600098, "genuineness": 3.873466968536377, "identity": 0.5939411520957947, "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_AUTH_reg/5b6db489d2019fc9bb480778b514ef7fa5cdff6318deb91886eb438fae5eb8f5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "5c3003988370c694f668d9112a9eaf4b213b9602dead4541882ca6777ccf5363", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "194c6f88da3676937c53008868021222cbfef9fd7f37f7374e41bfb3a94c4329", "generation_seconds_batch": 27.508233027998358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 5.824047088623047, "genuineness": 3.8169827461242676, "identity": 0.6809943318367004, "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_AUTH_reg/5c3003988370c694f668d9112a9eaf4b213b9602dead4541882ca6777ccf5363.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "5c43ac85aa15c1872d497c64b5691453e09dbb0bf7e06761008b926348035783", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe1ce4dded7f44b5f2ef46bc4e30f9270e9eb258f74fb2aaa868ad0d36c05ea5", "generation_seconds_batch": 20.650356878060848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 6.086154937744141, "genuineness": 4.036643981933594, "identity": 0.7269638180732727, "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_AUTH_reg/5c43ac85aa15c1872d497c64b5691453e09dbb0bf7e06761008b926348035783.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "5cbb16910f670b779a5a7ef757c63fe2e3378145825fdb70e188788d855b6a77", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0efb70bca1d32f24d60633441b885b294ee12912ebb22a377c00b4be7108e910", "generation_seconds_batch": 28.671361761982553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.555373668670654, "target_z": 3.2187497884329903, "ce": 6.197352409362793, "genuineness": 4.111956596374512, "identity": 0.43910589814186096, "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_AUTH_reg/5cbb16910f670b779a5a7ef757c63fe2e3378145825fdb70e188788d855b6a77.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": "5cca65600fab68dc6a39a03603f5af7365406343c854c1a8fcc5ab2b14813c5b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0c50b65bcd087b9b149779df23f4493603fab608b0e54cbdda0cd4a6ab498ff", "generation_seconds_batch": 19.739179227966815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.41180419921875, "genuineness": 3.8922953605651855, "identity": 0.36665573716163635, "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_AUTH_reg/5cca65600fab68dc6a39a03603f5af7365406343c854c1a8fcc5ab2b14813c5b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "5d107e477693ca94563d02bb5641cf0fe94a815346a1313807623a0be300db3f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "749c273f83ae328eb45f76b6eefda72e318c0c2433f74967740d9dea3de3ca05", "generation_seconds_batch": 25.89878777100239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.181467056274414, "genuineness": 4.344170093536377, "identity": 0.7232856750488281, "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_AUTH_reg/5d107e477693ca94563d02bb5641cf0fe94a815346a1313807623a0be300db3f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "5d35da4a27d9105496937cabf0d9edd3a148cca8b6e4d6ff9f44c09f709032e6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d66f9ed21f4931505725bb793f5387f753bf7f7ffd34ce41802032e50abe30b", "generation_seconds_batch": 24.797879952937365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402869701385498, "target_z": 2.1093751237129124, "ce": 6.340320587158203, "genuineness": 4.306513786315918, "identity": 0.6301554441452026, "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_AUTH_reg/5d35da4a27d9105496937cabf0d9edd3a148cca8b6e4d6ff9f44c09f709032e6.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "8e7fee78e3858bad8465c0006d5229101b572dcaf735f23f425393712c65deef", "generation_seconds_batch": 18.812717793043703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.575014352798462, "target_z": 1.312499992828237, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.673682689666748, "hypothesis": "La route tourna à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.05, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "5e39b96bdf9cd6551e9b7a24624c20d200765a07ca42d1c834808751b9da04a8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f17c38e5de7e470dc4d3eec42fc2d0df4d83268a6eef44bd381d1783f9e3c28", "generation_seconds_batch": 26.70499136694707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.356205940246582, "genuineness": 4.256305694580078, "identity": 0.43465471267700195, "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_AUTH_reg/5e39b96bdf9cd6551e9b7a24624c20d200765a07ca42d1c834808751b9da04a8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "5e9e93fdd8527e2aec585b7818def99dfae88d9c12e1d87aff2248fb55cdb38f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bde89570daa8b478988fe17c421e18d5d127af9e74949ea479840194b783111", "generation_seconds_batch": 24.9501206909772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8022687435150146, "target_z": 1.5312500107576446, "ce": 6.229123115539551, "genuineness": 4.105680465698242, "identity": 0.3855384290218353, "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_AUTH_reg/5e9e93fdd8527e2aec585b7818def99dfae88d9c12e1d87aff2248fb55cdb38f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "5f84548aa9fb44d11798d838ea3e1705c3797631f60083ba980841c3611d9d3e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d97888d1d61b9519fb1be362ef84f66936dbf5482e3c7ba01ace220edc03e80", "generation_seconds_batch": 25.126038472983055, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.571606159210205, "target_z": 3.234374822498865, "ce": 6.229123115539551, "genuineness": 4.074300289154053, "identity": 0.42709416151046753, "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_AUTH_reg/5f84548aa9fb44d11798d838ea3e1705c3797631f60083ba980841c3611d9d3e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "5fc2f86cc2fdd2982d62b9e27d2b77ef8e7a6f4f561b23536faa8fcbd6a5fa06", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62667604ff267c5864c9fd6a3b99891778510d4b81f0bf0915f927e8c4053a3d", "generation_seconds_batch": 24.56267498107627, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.745455265045166, "target_z": 1.4765626210235012, "ce": 6.268836498260498, "genuineness": 4.293961524963379, "identity": 0.714213490486145, "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_AUTH_reg/5fc2f86cc2fdd2982d62b9e27d2b77ef8e7a6f4f561b23536faa8fcbd6a5fa06.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": "5ff3b28853e48a88df55e1f2b98dd88f4c941dc9b5cba122f90a1c9f2e19d1a3", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d9dc5644a381c35699faf91ae2ef89aea36a33d2b359d1b1d0c81272f83c9d3", "generation_seconds_batch": 26.237925509922206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.070269584655762, "genuineness": 4.18099308013916, "identity": 0.6102335453033447, "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_AUTH_reg/5ff3b28853e48a88df55e1f2b98dd88f4c941dc9b5cba122f90a1c9f2e19d1a3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "60064e04ed6b3b30618f6c00becc02cea1395ccc6a49d29f0704b7f732ee0a27", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfa5779aea8ea8946ab96e49adf2e3010128a2aff649e1764191e638e9d84d09", "generation_seconds_batch": 25.537519666017033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.046441555023193, "genuineness": 3.9550557136535645, "identity": 0.5149803161621094, "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_AUTH_reg/60064e04ed6b3b30618f6c00becc02cea1395ccc6a49d29f0704b7f732ee0a27.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "607732319b377f33cab70f40a28b2e8f5cbac628e14f2705727acc21df4b95e8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11c5bed2ea24e1ff7762057b963bf16897280553ae44792910fb25dc5f704b24", "generation_seconds_batch": 27.72684156289324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.1894097328186035, "genuineness": 4.268857479095459, "identity": 0.4509690999984741, "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_AUTH_reg/607732319b377f33cab70f40a28b2e8f5cbac628e14f2705727acc21df4b95e8.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": "60a2b7958345c79c3d6822342c16df1e7d43412138573687c0715f73d3a796ff", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3867c8bdcd09a02d119afeff8c29fab76f06819c8f7ddece680d59b78faa2cef", "generation_seconds_batch": 22.47037289501168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8573784828186035, "target_z": 2.5468751595717274, "ce": 6.38797664642334, "genuineness": 3.8546390533447266, "identity": 0.38257867097854614, "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_AUTH_reg/60a2b7958345c79c3d6822342c16df1e7d43412138573687c0715f73d3a796ff.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "612a4571a21def2a458b93a9828e598641dd95e681eac9d2aca6fbbd948d3b8b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "662ea44bf41967ff95e309188d9fa23338e93423b334eb20ff2c4b3e63e646a3", "generation_seconds_batch": 26.094968223012984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.181467056274414, "genuineness": 4.13078498840332, "identity": 0.5742835402488708, "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_AUTH_reg/612a4571a21def2a458b93a9828e598641dd95e681eac9d2aca6fbbd948d3b8b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "618839560603d0fddf384eae5c20f1a26bb0b6749bd72b241083b679d9f6abb9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "38b66f17fc535d6dc61e3c873251414f7e758dfa3a7bd4d6e7f44a7bf481a4b1", "generation_seconds_batch": 25.940952660981566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9483609199523926, "target_z": 1.671875087854097, "ce": 6.213237762451172, "genuineness": 4.20609712600708, "identity": 0.25568732619285583, "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_AUTH_reg/618839560603d0fddf384eae5c20f1a26bb0b6749bd72b241083b679d9f6abb9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "61e30241b29a49991923e34634fc897ae04edd022f1e9e653db65628e11703ed", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e3445945377f7571e85b45c4a707b4a8a880cbd16a991bdbf38cbdec59b2355", "generation_seconds_batch": 26.693047498934902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.078212261199951, "genuineness": 4.21237325668335, "identity": 0.6639326810836792, "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_AUTH_reg/61e30241b29a49991923e34634fc897ae04edd022f1e9e653db65628e11703ed.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "6263b5500baae0c03d946ca170264b18cb30044396878846f322c91abffc5cd1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "061b29bcd315a8655c9dd61536f74b99f408aabfc3ebab619284b249a1b99084", "generation_seconds_batch": 25.36220547207631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.534433364868164, "target_z": 1.273437637159968, "ce": 6.332377910614014, "genuineness": 4.231201171875, "identity": 0.41055387258529663, "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_AUTH_reg/6263b5500baae0c03d946ca170264b18cb30044396878846f322c91abffc5cd1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "62688c80f1eea7518151f1e447895f7fdf64c9dd86e06fb154d9e9cf20198fbb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c5fa47bf1574d2e15da1bd9852c416fd7a15e7007a2a4ff085a769951a0f917", "generation_seconds_batch": 19.348872640985064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.0146708488464355, "genuineness": 4.042920112609863, "identity": 0.5356097221374512, "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_AUTH_reg/62688c80f1eea7518151f1e447895f7fdf64c9dd86e06fb154d9e9cf20198fbb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "629d79ea2305e90f06cd8ff3e6a3525a7c80c5ae06746e84c2257f62ae850ebf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a19431ef5141f512d1fc27b848342a52e89ec2e6e2158f2ebd2bbdb6de1ecd80", "generation_seconds_batch": 25.331858919002116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.165581703186035, "genuineness": 4.3127899169921875, "identity": 0.49614885449409485, "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_AUTH_reg/629d79ea2305e90f06cd8ff3e6a3525a7c80c5ae06746e84c2257f62ae850ebf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "641a609d04f5114a99258b93e393932e712622d7c28c846215ea3f588854e41a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e9da41a5dfea95fc40293fd4ee09a8eac4ac82681b1fef56034deb396600d218", "generation_seconds_batch": 25.7473061659839, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.045755386352539, "target_z": 1.7656248332565094, "ce": 6.062326908111572, "genuineness": 4.306513786315918, "identity": 0.7254339456558228, "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_AUTH_reg/641a609d04f5114a99258b93e393932e712622d7c28c846215ea3f588854e41a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "641f09fa3b86805b2514ddc04708ee2262ba95ce3e606cb1e27d502a954579f8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4860d2fef76c551fc1d07dc724508ebbeccbb5de6690f6a68ed3c68de6b66447", "generation_seconds_batch": 28.828807416022755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.467403411865234, "genuineness": 4.582659721374512, "identity": 0.7094112634658813, "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_AUTH_reg/641f09fa3b86805b2514ddc04708ee2262ba95ce3e606cb1e27d502a954579f8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "65c4c2b60bf71736562ea8f43bdecb7fdc7bef0fabdef8c8c1d336aac27a5fcc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f42ccb8607613f924526a0864eeb03a66241a966d16283e346b6d4b3d78b9017", "generation_seconds_batch": 24.52841027698014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.41180419921875, "genuineness": 4.13078498840332, "identity": 0.31414222717285156, "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_AUTH_reg/65c4c2b60bf71736562ea8f43bdecb7fdc7bef0fabdef8c8c1d336aac27a5fcc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "65d766cab21cf7b122677d5413c06a74ee55ea870640dfe5b3825812042c8f39", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0148f6a726ead8b118637cba92c5ca73d58b73d2296de6247d4b81579d607b7", "generation_seconds_batch": 25.05696232803166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4353346824645996, "target_z": 2.140625191844661, "ce": 5.589737892150879, "genuineness": 3.7793264389038086, "identity": 0.38072797656059265, "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_AUTH_reg/65d766cab21cf7b122677d5413c06a74ee55ea870640dfe5b3825812042c8f39.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "65fcb3322984921a36350fa42af02519269a5d0b59f45113937a7cdaba4667f8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49495c8de921e85427dffd79c69ef9b4365bd0b5661a1186bd9449c040bd35b6", "generation_seconds_batch": 29.06020836799871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.316492557525635, "genuineness": 4.074300289154053, "identity": 0.5921387672424316, "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_AUTH_reg/65fcb3322984921a36350fa42af02519269a5d0b59f45113937a7cdaba4667f8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "662d7a2e1baa735af8247f4b517ecc2cc3c3a8d1955fdd580eb8a57d78b10c94", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b95e8f8344e75871f85ee267c7577747c5d4f683514ac75ccb41a24677b1276", "generation_seconds_batch": 26.264507220941596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.205295085906982, "genuineness": 4.093128204345703, "identity": 0.5726794004440308, "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_AUTH_reg/662d7a2e1baa735af8247f4b517ecc2cc3c3a8d1955fdd580eb8a57d78b10c94.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": "669baed2eb990391742502e89b043c1d615e15a3ec52b21f5c77bbdd77d0c8b1", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c9f451a55a0b99dd0505d4b61314d92ee4aac2a19babc26861289df457e79647", "generation_seconds_batch": 16.378719346947037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.284721851348877, "genuineness": 4.143336772918701, "identity": 0.7363050580024719, "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": 0.03999999999999915, "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_AUTH_reg/669baed2eb990391742502e89b043c1d615e15a3ec52b21f5c77bbdd77d0c8b1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "66bdf01dbe899b5225e740181efd7c4f5646acdc96197636358809dc84e7b94d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c0769cca2c587cb8ffb7a57a5829694157cfb0056872245f13dc0348a13665d", "generation_seconds_batch": 25.16909567802213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.412689685821533, "target_z": 1.15624988166591, "ce": 6.41180419921875, "genuineness": 4.293961524963379, "identity": 0.3018350601196289, "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_AUTH_reg/66bdf01dbe899b5225e740181efd7c4f5646acdc96197636358809dc84e7b94d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "66ddcf3bf8c99ea4a51e5697e74672c0d8a2622a0b5c2fa9b3ab86554cb97cfd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37c260af181076609a377d166724c4b2981a1c9c50593fbb33cc079968e9bea3", "generation_seconds_batch": 27.879800538066775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.181467056274414, "genuineness": 4.557555198669434, "identity": 0.5147972702980042, "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_AUTH_reg/66ddcf3bf8c99ea4a51e5697e74672c0d8a2622a0b5c2fa9b3ab86554cb97cfd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "6721d9141d80c178b2e1343f8db0ba4187d7f12c8df09db373f03c4582af6e39", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cb4c45d36f2e95e1bd6e3ab5f06ba6f016a6c1e96abea29944276d8be3ded8db", "generation_seconds_batch": 25.472863759030588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.152970552444458, "target_z": 0.9062500251011706, "ce": 6.332377910614014, "genuineness": 4.21237325668335, "identity": 0.745385468006134, "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_AUTH_reg/6721d9141d80c178b2e1343f8db0ba4187d7f12c8df09db373f03c4582af6e39.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "6722826897a607028c1a732619477f383cc3a55b2cd4e114c52b103402ed9ca1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b5cc547a9d0e9825cac596233f311caeb472cd3fbde01a8d041b5aab583c90d", "generation_seconds_batch": 25.233970609027892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.875314712524414, "target_z": 1.6015624345576622, "ce": 6.157639026641846, "genuineness": 3.961331605911255, "identity": 0.5963684916496277, "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_AUTH_reg/6722826897a607028c1a732619477f383cc3a55b2cd4e114c52b103402ed9ca1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "673f2c3ea9630914b80bad644e8359fc28cbe6889462364dc08dfc91d67071cb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3c66e39103241fda499c7e924bd0ec0e7468e3087c9fdc19985dfd2fa94ce63", "generation_seconds_batch": 20.250342253013514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 6.41180419921875, "genuineness": 4.256305694580078, "identity": 0.29310208559036255, "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_AUTH_reg/673f2c3ea9630914b80bad644e8359fc28cbe6889462364dc08dfc91d67071cb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "6784d41c04edc670e71b8151dfba321d3242c4dcf1fe9304ad2d1e8d553422db", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7eba7bfe119ba3f0b75ce506f6f24eca29e52317daf20df84879636660f20e3b", "generation_seconds_batch": 26.08336986694485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019702911376953, "target_z": 2.7031250412376373, "ce": 6.332377910614014, "genuineness": 4.231201171875, "identity": 0.6465958952903748, "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_AUTH_reg/6784d41c04edc670e71b8151dfba321d3242c4dcf1fe9304ad2d1e8d553422db.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "6901deeb18db45cc4a69c395dbfe58bdea2a24feb98595702e33fd569b407e12", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df1977284d4fd7cbf07ce0d1d498a147f224bc762522a0ddabbc08a2545b924f", "generation_seconds_batch": 24.894474033964798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.054384231567383, "genuineness": 4.293961524963379, "identity": 0.5355159640312195, "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_AUTH_reg/6901deeb18db45cc4a69c395dbfe58bdea2a24feb98595702e33fd569b407e12.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "69a8cb46b4d249d7bfef3a856c267ab4465dd753681585a5918910b888cacd6b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "75fd16a408af9ef4da196b5ae019042bbb90955ddc857794c7274d12b2bfbbc2", "generation_seconds_batch": 28.93567006511148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.889843463897705, "target_z": 2.5781252277034765, "ce": 6.348263263702393, "genuineness": 4.049196243286133, "identity": 0.5359801054000854, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem Altendfenster. Wir können gemeinsam hineinschreiten.", "wer": 0.1875, "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_AUTH_reg/69a8cb46b4d249d7bfef3a856c267ab4465dd753681585a5918910b888cacd6b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "69b04f96ecf3f139e788eafb42a74f8b09d2b83dece61d76fa59d9c0bb824d70", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81ecbd1ad9a886a4e1dadf0c5d9d65839cd66d480e1bff72f9c6842eb71097b3", "generation_seconds_batch": 25.586016373010352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.38797664642334, "genuineness": 3.911123275756836, "identity": 0.5088740587234497, "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_AUTH_reg/69b04f96ecf3f139e788eafb42a74f8b09d2b83dece61d76fa59d9c0bb824d70.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "69e789a3de81cb9c59efdb83b2c4622032b80756b103858cdd4095756898f276", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eedadc6d97ab73bf2a14c50907e8139c47542dfb127aa9d5dabf979bc8f48ead", "generation_seconds_batch": 19.236081793089397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9077796936035156, "target_z": 1.6328125026894111, "ce": 6.252951145172119, "genuineness": 4.055472373962402, "identity": 0.32101404666900635, "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_AUTH_reg/69e789a3de81cb9c59efdb83b2c4622032b80756b103858cdd4095756898f276.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "6a25fc9e0825213fa5f0a252b2a8543f57370cf5556a66b44c86c5443b54215d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "662188f8397f068cd18d5a985249274fe2558d4087e97f8a38abdc032ec9a2b3", "generation_seconds_batch": 26.923399708000943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.4830043315887451, "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_AUTH_reg/6a25fc9e0825213fa5f0a252b2a8543f57370cf5556a66b44c86c5443b54215d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "6a5c80229a6af7940683ec0c85cd50f20fadea06b15766491fa3e4ca48fc5aef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "17ca48a5467e5f62c91d4c4c43badc2fb85e3718c7c99e66c12d8ea41bc5def2", "generation_seconds_batch": 24.556657046079636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 6.165581703186035, "genuineness": 4.256305694580078, "identity": 0.7176755666732788, "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_AUTH_reg/6a5c80229a6af7940683ec0c85cd50f20fadea06b15766491fa3e4ca48fc5aef.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "464399843fcb9e47598b8b994ec0674e9b7c2aa4880451ec80d027855b037eea", "generation_seconds_batch": 17.923304528929293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9077796936035156, "target_z": 1.6328125026894111, "ce": 6.316492557525635, "genuineness": 4.017816066741943, "identity": 0.2398516833782196, "hypothesis": "I wrote the number on that envelope, read it carefully, and check the last two digits. Call me when you have a moment.", "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_AUTH_reg/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "6a81bc06ed452d2aa7e9ecf177e493262b8ffff7c7760da1c766248c0602956d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b81b81af00d3a122260568ddc0a0c2c4f22cbcb3a38515014a5c6c986ace49b", "generation_seconds_batch": 25.23112268501427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1350340843200684, "target_z": 1.8515625206188187, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.5481353998184204, "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_AUTH_reg/6a81bc06ed452d2aa7e9ecf177e493262b8ffff7c7760da1c766248c0602956d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "6a885b5857c8946fb45f4163a05ca381eff56d1e6b128c74168205b780028d4c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40d46f1c012829b9d3fea06f47c43bd1c9d154680ed9ece5ddaf7926e90e703f", "generation_seconds_batch": 20.28801698004827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 5.990842819213867, "genuineness": 4.017816066741943, "identity": 0.5551663041114807, "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_AUTH_reg/6a885b5857c8946fb45f4163a05ca381eff56d1e6b128c74168205b780028d4c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "6aa0c5a119df512994f6f504d4d377b2f5614d37cde99ab853de79d6bbfedf8c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a278795372ecc03a3e44ea1dc72fd42bde1685fb486cd14acfc10543890d51b", "generation_seconds_batch": 25.85885436506942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9889419078826904, "target_z": 1.710937443522366, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.5310443639755249, "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_AUTH_reg/6aa0c5a119df512994f6f504d4d377b2f5614d37cde99ab853de79d6bbfedf8c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "6ab909c4ca0920485a00fab36064ee68cb8c35fb07968fe51919181249326463", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ac816b8aef51fbd8e212bfadb03bf6c1b8a74db8155a4a20678bf1bc712d0f5", "generation_seconds_batch": 26.133559937006794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.141753673553467, "genuineness": 4.017816066741943, "identity": 0.5785682797431946, "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_AUTH_reg/6ab909c4ca0920485a00fab36064ee68cb8c35fb07968fe51919181249326463.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "6ace3e149842513e9eeddca561e1cc54f9d777fc2bc7da60c94490c8cccb1d5a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "faea5f9d3a17f32f188cb75c757fb4e41d7770efe12e7f8e562425241bcb04c7", "generation_seconds_batch": 18.682038645027205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0944528579711914, "target_z": 1.8124999354541327, "ce": 6.070269584655762, "genuineness": 4.632867813110352, "identity": 0.6600576639175415, "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_AUTH_reg/6ace3e149842513e9eeddca561e1cc54f9d777fc2bc7da60c94490c8cccb1d5a.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": "6adf2db8f68fa755a9b9238b0f10ae3df9ac0d050301bd767afe518f36b586ac", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3bfc60e215002d8b2c059eace953e33a3aa00f55920b95217adbc5a9396d946", "generation_seconds_batch": 25.697462563053705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.38003396987915, "genuineness": 4.055472373962402, "identity": 0.7330008745193481, "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_AUTH_reg/6adf2db8f68fa755a9b9238b0f10ae3df9ac0d050301bd767afe518f36b586ac.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "22eba58429eeaca1448ab226b5a572f8158ec58f35dc0a2dae82fe582c2b9b32", "generation_seconds_batch": 18.605552700930275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.169203042984009, "target_z": 0.921875059167045, "ce": 6.157639026641846, "genuineness": 4.68307638168335, "identity": 0.8103958368301392, "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_AUTH_reg/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "6ba79ea45f48e797dc9ee389fcc47d07809785fa1dde66f5d95c06e5b4be817f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "336a24809d23640d344e4df9d868095977812853c634dabcb7b7b5a08434e24f", "generation_seconds_batch": 20.650356878060848, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.5891746282577515, "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_AUTH_reg/6ba79ea45f48e797dc9ee389fcc47d07809785fa1dde66f5d95c06e5b4be817f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "6bc62c0778dacbae7b5b81f84bfe376274cb4b9d8eee5cf9858082b1eab5835b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa8d8dd3ac4948903fa051d4db3799fd980de692db47094789512771ac03607a", "generation_seconds_batch": 26.49455736298114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.197352409362793, "genuineness": 4.231201171875, "identity": 0.6393471360206604, "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_AUTH_reg/6bc62c0778dacbae7b5b81f84bfe376274cb4b9d8eee5cf9858082b1eab5835b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "6c1399e23308159396042786e98d4d74b746dc631e3ca364cafc247e074fc25c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "285ba92fae38d1ed2a6e865f6f891566a68f46ebc00c34afd2f0f5cc6be6a8dc", "generation_seconds_batch": 24.176364543032832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0538716316223145, "target_z": 1.7734373502894467, "ce": 6.41180419921875, "genuineness": 4.356721878051758, "identity": 0.665284276008606, "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_AUTH_reg/6c1399e23308159396042786e98d4d74b746dc631e3ca364cafc247e074fc25c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "6c574de259b748055a7efd35d8f415daebfab50e70580df7206b5d22b8b87f3e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2bf8b3f69b555f2d08a47dd573ea64a79813c0a5ef88ce433c2a4a4b836e315c", "generation_seconds_batch": 24.935790853924118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.062326908111572, "genuineness": 4.005263805389404, "identity": 0.5019497275352478, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "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_AUTH_reg/6c574de259b748055a7efd35d8f415daebfab50e70580df7206b5d22b8b87f3e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "6c9707933b5c5a117b1b57289ec6cc8c6c93b43dd4495260d52749b0c448da74", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "77880a4c305f80ac6efbc6a51da8f4491e1874a2d64158e0033797aa81fb4665", "generation_seconds_batch": 26.20526841294486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.324435234069824, "genuineness": 4.256305694580078, "identity": 0.31127578020095825, "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_AUTH_reg/6c9707933b5c5a117b1b57289ec6cc8c6c93b43dd4495260d52749b0c448da74.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "6cf353326dedc257181df4bd01c4caf1aaec1bc614e3287b7fc30c837588fc78", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e2717ac1afe111bc051138f816aab2a469d7363574bf8acf8c95e38acb41413", "generation_seconds_batch": 25.80701417394448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.117925643920898, "genuineness": 4.20609712600708, "identity": 0.48836272954940796, "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_AUTH_reg/6cf353326dedc257181df4bd01c4caf1aaec1bc614e3287b7fc30c837588fc78.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "6d04fd2a719064a92aedb84b37cd0304208f20195a8a4570dd2bb9ba22190c93", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0e7e899728a18e3bbaa0830a7a3db28d9d5d2e2d7e6e9f4bc93059e7a92e17d", "generation_seconds_batch": 18.518168642069213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.459460258483887, "genuineness": 4.469690799713135, "identity": 0.6101429462432861, "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_AUTH_reg/6d04fd2a719064a92aedb84b37cd0304208f20195a8a4570dd2bb9ba22190c93.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "6d908bc4ad2493920909592e2543aef78b895eee5af73a4dd38614ad02702a91", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d14d7c55408cc18c258c5a2954f2f8730a64af66ca6d7b39789d22bb0910568", "generation_seconds_batch": 25.603032885934226, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.133810997009277, "genuineness": 4.030367851257324, "identity": 0.5961413383483887, "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_AUTH_reg/6d908bc4ad2493920909592e2543aef78b895eee5af73a4dd38614ad02702a91.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "6dd0a3728e356e8da5e4eb8237149949e9fe6034a8980c0ebfd0520d3de325a1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6ed5404f88535c5bb8d2ad5889ed1cce1923ab43b7cc253f378c48076172dea", "generation_seconds_batch": 25.23112268501427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.229123115539551, "genuineness": 4.406930446624756, "identity": 0.4735175669193268, "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_AUTH_reg/6dd0a3728e356e8da5e4eb8237149949e9fe6034a8980c0ebfd0520d3de325a1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "6dd854724ed0a2a87f0861ef5c2eeba6c55eb4ff821a6c19762bc555c894a1de", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e962c294cad597e5b8807b7ec1c994ede76fa94e905a83111e75eab87430d998", "generation_seconds_batch": 25.29584223101847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.759983539581299, "target_z": 2.453124955176481, "ce": 6.395919322967529, "genuineness": 4.042920112609863, "identity": 0.5931963324546814, "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_AUTH_reg/6dd854724ed0a2a87f0861ef5c2eeba6c55eb4ff821a6c19762bc555c894a1de.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "6df24d25babdfa5400a702fc4600bb8d3f0ac12eec2aabdafeb2a591a7aebfc6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "747aa4847092b3d4dd874adde5cbd07657b2d8c72343f69c2df301eaf71113d7", "generation_seconds_batch": 20.35694746603258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.435632705688477, "genuineness": 4.243753433227539, "identity": 0.326535701751709, "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_AUTH_reg/6df24d25babdfa5400a702fc4600bb8d3f0ac12eec2aabdafeb2a591a7aebfc6.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": "6df43d74d8e341c2d3abe3276d7a01fa6b7722dc126396a285d9fd6e8090be2d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "239ed9eb57a8582df16b9fde6154ac089b1629328bdd6b963f0e978c18e9a539", "generation_seconds_batch": 22.016084103961475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.117925643920898, "genuineness": 4.18726921081543, "identity": 0.7044597268104553, "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_AUTH_reg/6df43d74d8e341c2d3abe3276d7a01fa6b7722dc126396a285d9fd6e8090be2d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "6e4259d877d1f887a96f701359fc5410c723f5c3cc33ea829a80fbe1fe61b1e9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a725b5c83c1e03bb35c1e3e582a1d4bed1f944df57937929d407da64971b75c1", "generation_seconds_batch": 25.411230052937753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 5.95112943649292, "genuineness": 4.243753433227539, "identity": 0.6845275163650513, "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_AUTH_reg/6e4259d877d1f887a96f701359fc5410c723f5c3cc33ea829a80fbe1fe61b1e9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "6f0ce3f180dcc087c48a71d26c5494c94114afea40c0f1e9ad666dd6ed3f3fe5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e4d7f0844201d448b489f01085ea2226b08046db6dcbe32b1f7bdfc15efc260", "generation_seconds_batch": 29.309814302017912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.339643716812134, "target_z": 1.0859374578658922, "ce": 6.395919322967529, "genuineness": 4.3881025314331055, "identity": 0.4403975307941437, "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_AUTH_reg/6f0ce3f180dcc087c48a71d26c5494c94114afea40c0f1e9ad666dd6ed3f3fe5.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": "6f6cddaa2508488d2464e9abfe52c798a39dc2b2665680aae09724b53d515af8", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b74147d9542c16c04a09aead316c4a5349ef0f76f131c09e25042f12aa8c5ec1", "generation_seconds_batch": 22.435941111063585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.348263263702393, "genuineness": 3.8797430992126465, "identity": 0.36345165967941284, "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_AUTH_reg/6f6cddaa2508488d2464e9abfe52c798a39dc2b2665680aae09724b53d515af8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "6f74bc13e309cd1751551007c8806e1b68e7fd67c41502aa6af8eb69cfbef330", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "895c744cc2d3af4063012e367dc15eb75047cdeffb99c08e397c29e85eba27cc", "generation_seconds_batch": 27.70692950708326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.09409761428833, "genuineness": 4.3127899169921875, "identity": 0.7998997569084167, "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_AUTH_reg/6f74bc13e309cd1751551007c8806e1b68e7fd67c41502aa6af8eb69cfbef330.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "6f74c2a2b6cf698c0ec20932e9edf3295aae81c65c2134d1de5ac66119166bcc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab11cceb967d572c2360bb74fb9199a2e0e4f2f21fa147f662de67472457e6b1", "generation_seconds_batch": 26.245415218989365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1042730808258057, "target_z": 0.8593749229035473, "ce": 6.3641486167907715, "genuineness": 4.557555198669434, "identity": 0.3943108022212982, "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_AUTH_reg/6f74c2a2b6cf698c0ec20932e9edf3295aae81c65c2134d1de5ac66119166bcc.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": "6fd2c7d45e9487bf90dab54521bbb17b3887f501c16c70a4779c5f10b388014c", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f1b60dfa77fbdd854fde1b97e5ab453ff236f429ee2402c3b4aab9951c60caa", "generation_seconds_batch": 22.15884544304572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.205295085906982, "genuineness": 4.256305694580078, "identity": 0.5354238748550415, "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_AUTH_reg/6fd2c7d45e9487bf90dab54521bbb17b3887f501c16c70a4779c5f10b388014c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "7037312f45f091b50ca13ce202050c8e5c3245b28dc83b88ff904f3007a418b9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28c921838827df3a7dd03bf7e5a28eb07b6feec000bb64f88fb33fbd90dcb0da", "generation_seconds_batch": 25.554971315083094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.5064305067062378, "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_AUTH_reg/7037312f45f091b50ca13ce202050c8e5c3245b28dc83b88ff904f3007a418b9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "705b86b70b9d16716596f88cfd0d56ea44edea711cc2792989678b96b8ef766f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11d1fc7cf39fd72591f8ed03fd0635312f488545a7e168c311ea46e1a1a718d6", "generation_seconds_batch": 27.393418796011247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.046441555023193, "genuineness": 4.362998008728027, "identity": 0.6919316053390503, "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_AUTH_reg/705b86b70b9d16716596f88cfd0d56ea44edea711cc2792989678b96b8ef766f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "706bf62b7bf4b3d9d2103f8b2dd7a1ac27abf4e233cdc1b25f9244b9592452d2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5441fe40f0704da412f8327af3b0be04fcec603887e6b73df8aca5e56fa7c03", "generation_seconds_batch": 25.204119879985228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.510084629058838, "target_z": 1.2500000860611564, "ce": 6.070269584655762, "genuineness": 4.419482707977295, "identity": 0.7565129995346069, "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_AUTH_reg/706bf62b7bf4b3d9d2103f8b2dd7a1ac27abf4e233cdc1b25f9244b9592452d2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "706c8d5b921d84f7e858e9212ed8782419550e4c9caa211a4a2d1fe3800f67a8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3994ec0f5de6d75d2dfd608cece10ea245235c0cd4281ac473fe972d417a000f", "generation_seconds_batch": 27.49681421497371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9223079681396484, "target_z": 2.609374836842391, "ce": 6.205295085906982, "genuineness": 4.068024158477783, "identity": 0.48180091381073, "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_AUTH_reg/706c8d5b921d84f7e858e9212ed8782419550e4c9caa211a4a2d1fe3800f67a8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "70dbecc2e5a74fb0bf37a2b48e341d29c64e85ad04ca13d83aba25acb07e1f76", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66f2cf8b0f63a90271bbc34aba2951137888afdd9ae084cd995e489f19a03bec", "generation_seconds_batch": 18.461136849946342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5263171195983887, "target_z": 1.2656251201270308, "ce": 6.395919322967529, "genuineness": 4.319066047668457, "identity": 0.29071375727653503, "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_AUTH_reg/70dbecc2e5a74fb0bf37a2b48e341d29c64e85ad04ca13d83aba25acb07e1f76.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "70fef5934465f56b8b7572f1f91315965ec18c1e44f0cedd00880faa010ca751", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1464176162dca64ad1f328bf215aba3d175a653396be4617ae22e8c1e606cfb", "generation_seconds_batch": 25.82176985195838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.046441555023193, "genuineness": 4.494794845581055, "identity": 0.6934388875961304, "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_AUTH_reg/70fef5934465f56b8b7572f1f91315965ec18c1e44f0cedd00880faa010ca751.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "712f8a0a12c5d6421ee68529ae6e559b6089fe4f19443b65726d9cf96b32798b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "872abff7fb43baf318c68947be9e98e932034744e6882b2211eb34db3c84bd9d", "generation_seconds_batch": 18.711568597937003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 6.308549880981445, "genuineness": 4.143336772918701, "identity": 0.6362619996070862, "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_AUTH_reg/712f8a0a12c5d6421ee68529ae6e559b6089fe4f19443b65726d9cf96b32798b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "714af3c3c9564c865bdde22582393dbd688436d6da931150e7b29f122b42ea20", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c997648c8c88e51bf35e13f77bafdecb4e212810c0ae8889791e5e0ff53b995", "generation_seconds_batch": 27.05371477000881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.133810997009277, "genuineness": 4.030367851257324, "identity": 0.529784083366394, "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_AUTH_reg/714af3c3c9564c865bdde22582393dbd688436d6da931150e7b29f122b42ea20.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "714ce96729dac388ca662935835983ab6762b136246e5579a320d2fb5bb7fe55", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "82f8dc0d6e3becccd556f17c841b8b80fd31b746451aed245c3503cd097087c8", "generation_seconds_batch": 26.290578097919933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1448543071746826, "target_z": 0.8984375080682334, "ce": 6.41180419921875, "genuineness": 4.218649387359619, "identity": 0.3881191909313202, "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_AUTH_reg/714ce96729dac388ca662935835983ab6762b136246e5579a320d2fb5bb7fe55.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "718ea0d3c2f3a6e56f62c57222b552fdc6b34c60de844005ba19e345479135c2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b052973c94db5a55c8e236ad6f264374a15f0db97cf7627d21d7aee42dd9c3c", "generation_seconds_batch": 26.138799433945678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.284721851348877, "genuineness": 4.281409740447998, "identity": 0.5945825576782227, "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_AUTH_reg/718ea0d3c2f3a6e56f62c57222b552fdc6b34c60de844005ba19e345479135c2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "71a060af3bd9331e8fe4864890a84f449dbea89641c99d88174683f53eb5acfd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acfb82ccba346009793cdd73344c9bf230d35b53b937022b713867e841399b4c", "generation_seconds_batch": 26.245415218989365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8671984672546387, "target_z": 1.5937499175247252, "ce": 6.4515180587768555, "genuineness": 4.124508857727051, "identity": 0.4419125020503998, "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_AUTH_reg/71a060af3bd9331e8fe4864890a84f449dbea89641c99d88174683f53eb5acfd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "71a36a85538ea4c211e38deed1359eb9aebfdcd166128bdb84f7af2ee3f47b55", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e380de0eb30ed5ce5c9f40d05c60d30520e604f32d6bb94ca57040d66f5b949", "generation_seconds_batch": 19.730083729955368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7779200077056885, "target_z": 1.507812459658833, "ce": 6.252951145172119, "genuineness": 4.093128204345703, "identity": 0.27295976877212524, "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_AUTH_reg/71a36a85538ea4c211e38deed1359eb9aebfdcd166128bdb84f7af2ee3f47b55.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "71a3bd86317dc0bc566618bb2325f65fff9f25921253e55c127e53e154444af0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5465aed16c1212908c98e53333f5680d4baa2c317ac3efa8243e0f8262ff1e13", "generation_seconds_batch": 25.43126228894107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.09409761428833, "genuineness": 4.15588903427124, "identity": 0.7910482287406921, "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_AUTH_reg/71a3bd86317dc0bc566618bb2325f65fff9f25921253e55c127e53e154444af0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "7232936a45a9c1f8ef111c031dc2fd280782f54fdaa68156f453438f73932af3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f2c2f52ec67a7ad88c6e2a8f7254185d11dab73d5584c41eb9e45637e8c8fe6", "generation_seconds_batch": 26.290578097919933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.05963397026062, "target_z": 0.8164063087188098, "ce": 6.3641486167907715, "genuineness": 4.570107460021973, "identity": 0.38829416036605835, "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_AUTH_reg/7232936a45a9c1f8ef111c031dc2fd280782f54fdaa68156f453438f73932af3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "72e9120bcc5d7a4b222c07ba9ab5022109446ef456fdf151447fcbc04e03299c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b507ec92d3067e5f1add20aa8956399c8e2b00d986316464a540a6492765ebe", "generation_seconds_batch": 24.794424962019548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9622392654418945, "target_z": 0.7226563338199805, "ce": 6.459460258483887, "genuineness": 4.444586753845215, "identity": 0.48197412490844727, "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_AUTH_reg/72e9120bcc5d7a4b222c07ba9ab5022109446ef456fdf151447fcbc04e03299c.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": "73124664b0cc6b6a8ce2df7c60b32e60cbdcd442ba35d61c3dd61b50103e7c34", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79ddb4902f8bff72696b54e6f956d54a184e4c10fd6ab1a60d1b4906c5cfcf63", "generation_seconds_batch": 16.100515394937247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.443574905395508, "genuineness": 4.231201171875, "identity": 0.4284089207649231, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann lagte ich das Handy weg. Sag mir, wann wir reden.", "wer": 0.08333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/73124664b0cc6b6a8ce2df7c60b32e60cbdcd442ba35d61c3dd61b50103e7c34.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": "7327d0c18a56234db32e151121b039db262e76052503f2ce37352431e59bbe65", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcb38bc409129bbfae7b0eb4da3fde596e1516a3b1ea727837f60d3a1468bf9e", "generation_seconds_batch": 16.499336790060624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246957302093506, "target_z": 2.921875059167045, "ce": 6.41180419921875, "genuineness": 4.400654315948486, "identity": 0.3175758719444275, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "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_AUTH_reg/7327d0c18a56234db32e151121b039db262e76052503f2ce37352431e59bbe65.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "732a2002f34e5f944d51b4df8f90e09bd2e972c2cc42fe2f35fcbbc6a4ad03a7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7c56e382f9b835521a0768252578e108ffba09d1ab14692b280ddccdda87317", "generation_seconds_batch": 20.168883408070542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.117925643920898, "genuineness": 4.3127899169921875, "identity": 0.6674295663833618, "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_AUTH_reg/732a2002f34e5f944d51b4df8f90e09bd2e972c2cc42fe2f35fcbbc6a4ad03a7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "7349bb9f5d39399c3f28daa09f4dd01a716b2d54c5a36eb93d447b336e4b6e5a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9191ebcb81dd029edc069a963d3673e1fd216e13eba75b463886293e7d9c898b", "generation_seconds_batch": 27.341825028066523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2486610412597656, "target_z": 1.9609373000871053, "ce": 6.348263263702393, "genuineness": 4.218649387359619, "identity": 0.33909282088279724, "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_AUTH_reg/7349bb9f5d39399c3f28daa09f4dd01a716b2d54c5a36eb93d447b336e4b6e5a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "73736e76bb70cd2a83b4aa580b4e2abc73ae0635943415a04bc8fec920a1622a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cfa1a4643f041cf794b56afdf1e9f94c8f11c84d1d5660e96923f439ebb5868", "generation_seconds_batch": 25.731791501049884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.221180438995361, "genuineness": 4.331617832183838, "identity": 0.7280615568161011, "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_AUTH_reg/73736e76bb70cd2a83b4aa580b4e2abc73ae0635943415a04bc8fec920a1622a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "73f873e1ed47050229766ebc511934001da8669d15591c31320fc576ffd28933", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6f000b5a13395b879bea280ae4eb48e6bfb09540f3bdcfe08063a7af62e10f39", "generation_seconds_batch": 20.35694746603258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5888926982879639, "target_z": 0.3632812387941203, "ce": 6.252951145172119, "genuineness": 4.519899368286133, "identity": 0.5097295641899109, "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_AUTH_reg/73f873e1ed47050229766ebc511934001da8669d15591c31320fc576ffd28933.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "7401b662caa9159636bf6369b860d3c9e66e75225c9d76b1a1b1498934079d6b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53d0afb34fc6b053f144b3805ed884766ba93f85fbed6646856db8db8e37062d", "generation_seconds_batch": 20.443757848930545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.469503402709961, "target_z": 1.2109375008964705, "ce": 6.213237762451172, "genuineness": 4.632867813110352, "identity": 0.6269907355308533, "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_AUTH_reg/7401b662caa9159636bf6369b860d3c9e66e75225c9d76b1a1b1498934079d6b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "740ff2e79466cffd5e4ec1e96ec2fba89f5332f756409434dff8844f7ebc5ecf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d06fba3ab0616a959313b451eb53d4c4a5a954f4ed0d3bb2ae9371e30b02769d", "generation_seconds_batch": 25.60334236000199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.340320587158203, "genuineness": 4.061748504638672, "identity": 0.4588971734046936, "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_AUTH_reg/740ff2e79466cffd5e4ec1e96ec2fba89f5332f756409434dff8844f7ebc5ecf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "7431828f938374d397096e4a41858ebed91d9e92bbe38c35202e3947dd816c22", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c43aecbd6598233f83268bf7a49b661296691725696255fd55874351bc8577ea", "generation_seconds_batch": 26.382820946979336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.109982967376709, "genuineness": 4.18099308013916, "identity": 0.6105055212974548, "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_AUTH_reg/7431828f938374d397096e4a41858ebed91d9e92bbe38c35202e3947dd816c22.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "74a0b760eaadd8198f1360ad9750709f35bafe5b88bbaad9cfd9126b8d30cc17", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "936363b4603c3bb26b807c0ea00f5c27313f00740f982f4a10e65ff51e8d4502", "generation_seconds_batch": 25.358860118081793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13332986831665, "target_z": 2.812499820705924, "ce": 6.0146708488464355, "genuineness": 4.030367851257324, "identity": 0.6439365744590759, "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_AUTH_reg/74a0b760eaadd8198f1360ad9750709f35bafe5b88bbaad9cfd9126b8d30cc17.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "75578d52cb525e100a1d0d49c52ce0c3723e3d2b37e06a464a9ead32ddc4cf18", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70982f87d7c790e4fe51a47ddd18af64876be5b620a3156e862d038378805e7d", "generation_seconds_batch": 21.310535597032867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 5.696964263916016, "genuineness": 3.998987913131714, "identity": 0.6297677755355835, "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_AUTH_reg/75578d52cb525e100a1d0d49c52ce0c3723e3d2b37e06a464a9ead32ddc4cf18.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "756e727b0912adb51f1dc7924f6e752e13c40b2b69cbf742e1564e2e67cb90c9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "666d643540562cee1ae7a7e59d4436c2e8efef12c64d6003a30325317cea5d1e", "generation_seconds_batch": 26.15496964205522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 5.919358730316162, "genuineness": 4.300237655639648, "identity": 0.6721408367156982, "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_AUTH_reg/756e727b0912adb51f1dc7924f6e752e13c40b2b69cbf742e1564e2e67cb90c9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "75c418e7055200781555037f19358f19edc8799e281aa145e6667111ff8cf6d9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "808daeb01da1decfd1fbb1728114decd41d144b57c78cb9c386d6b53a022c3fe", "generation_seconds_batch": 26.86438907601405, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.435632705688477, "genuineness": 4.093128204345703, "identity": 0.5840343236923218, "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_AUTH_reg/75c418e7055200781555037f19358f19edc8799e281aa145e6667111ff8cf6d9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "7704d90eb58c29f52f183bae91fc7d44f3852234bb33608c5fce53a4eb755220", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dc1f7a0c72639683d0dc5898af5518dd0ae42f241530eeb1b431aee479dc125", "generation_seconds_batch": 24.811311046010815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.23706579208374, "genuineness": 3.8169827461242676, "identity": 0.20917944610118866, "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_AUTH_reg/7704d90eb58c29f52f183bae91fc7d44f3852234bb33608c5fce53a4eb755220.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "773e0e784b0806b8f07c7725c032d5e4f99d7eb82a714c57649a9efb4b20d29e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e5116bdc1597931f477e5525cf87c5a26a812030d0d430f52bfbf94973e42d", "generation_seconds_batch": 21.847837987937964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 5.982900142669678, "genuineness": 4.005263805389404, "identity": 0.5436728000640869, "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_AUTH_reg/773e0e784b0806b8f07c7725c032d5e4f99d7eb82a714c57649a9efb4b20d29e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "776ac2bf3c30f6fd7a8f619ef34a2d5454421e184883d1e7e9785d17df2ee5a7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4d2c8db6075ee17a8cef0251530cf02c0a29068e0f7fe019818ddf54fcfff5a", "generation_seconds_batch": 24.58940027898643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8428497314453125, "target_z": 1.5703123664259135, "ce": 6.157639026641846, "genuineness": 3.9801597595214844, "identity": 0.2806500792503357, "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_AUTH_reg/776ac2bf3c30f6fd7a8f619ef34a2d5454421e184883d1e7e9785d17df2ee5a7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "7794b3f46ab384d03c8fa4ff48907e9dfedfdfe65b41e0cb093c9e6dc60f5211", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73e212f10e9e87296873ebb51ada9610c1cbead0f0558bf26b011138d0e56441", "generation_seconds_batch": 22.802018410991877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.5676553249359131, "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_AUTH_reg/7794b3f46ab384d03c8fa4ff48907e9dfedfdfe65b41e0cb093c9e6dc60f5211.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "74d2dbd4bebf816fc81fd5c21ce460cce0c02ac15fb2f2ce42a1fd0b125f2dd9", "generation_seconds_batch": 18.483687986037694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.308549880981445, "genuineness": 4.469690799713135, "identity": 0.7372502684593201, "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_AUTH_reg/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "77b5f5043ac953049f81a47f5ceac2978b4c0636d142693130b788ea99c3c1f5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50d7bf1124e1d0f92f01d4e7c34e27df3f6d5806adc3e3be2b4b0748a890f78f", "generation_seconds_batch": 27.312456307001412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 5.728734970092773, "genuineness": 4.042920112609863, "identity": 0.7331769466400146, "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_AUTH_reg/77b5f5043ac953049f81a47f5ceac2978b4c0636d142693130b788ea99c3c1f5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "77c341b40b141f172af8fcf53b311a4a214a329424d979436e0cbbaf7410d429", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "64df28cea0f026add182111e3a1ae35f708cbd237632c1eb4fc6ab2e73daac63", "generation_seconds_batch": 25.331858919002116, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.09409761428833, "genuineness": 4.18099308013916, "identity": 0.6027024388313293, "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_AUTH_reg/77c341b40b141f172af8fcf53b311a4a214a329424d979436e0cbbaf7410d429.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "78181228c303c8fc7a94432d0cde68e303563513de772647cf0da49c42cd4b47", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54b7a1334ae2a5fa754929b4c0ee30923404d452bf2116a6e7241db5fbbb8a17", "generation_seconds_batch": 18.95389297895599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.6277990341186523, "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_AUTH_reg/78181228c303c8fc7a94432d0cde68e303563513de772647cf0da49c42cd4b47.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "78659ea1384a2530babfe93c6991e30de9e2ed6b0769253a371a6e509ff194a7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3f7c04a631598aa7645a7b09e6ca2d444e786ae8068975206efa3c7281995b14", "generation_seconds_batch": 17.965504886000417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.295654773712158, "target_z": 2.9687501613646683, "ce": 6.23706579208374, "genuineness": 4.168440818786621, "identity": 0.4517858028411865, "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_AUTH_reg/78659ea1384a2530babfe93c6991e30de9e2ed6b0769253a371a6e509ff194a7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "792b2917e1f5ed0b18e562a3844e95e5a743a61a6056dc3b2a69498cb74e1c92", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7084e916bb415fe835ecf3c9f7f9102cf1fd4b56e4c3a60566e7cc860b2d9d86", "generation_seconds_batch": 26.178430213942192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.558781862258911, "target_z": 1.2968749587623625, "ce": 6.284721851348877, "genuineness": 4.482243061065674, "identity": 0.7933112382888794, "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_AUTH_reg/792b2917e1f5ed0b18e562a3844e95e5a743a61a6056dc3b2a69498cb74e1c92.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "f729d6819a5afd61269885df228c6477a7b8621e8716da7e65dd4a4ff8a23c07", "generation_seconds_batch": 20.008315172046423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.260893821716309, "genuineness": 4.268857479095459, "identity": 0.7882985472679138, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de laag en daarna naast het raam, neem ze mee als je weggaat.", "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_AUTH_reg/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "79b8b304586c73398cd0ef6c96af933a6bda8490d4817af95c5cee5bd0e9d9fe", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60f79b264294cfb3ce7be1d1e9a8b5e7748b74e56790eaf156d092d697688fea", "generation_seconds_batch": 25.971301213023253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.125868320465088, "genuineness": 3.6663577556610107, "identity": 0.4813576638698578, "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_AUTH_reg/79b8b304586c73398cd0ef6c96af933a6bda8490d4817af95c5cee5bd0e9d9fe.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "7a0ef428ea7169c99eae0f28c987bd87c2863d703e34be52c6e87daca80687a1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bec9262e8a4bca31a522f445f605522aaea4bfdb31815ce1b4d59b13de276a42", "generation_seconds_batch": 19.040464148973115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.063692092895508, "target_z": 0.8203125672352785, "ce": 6.4515180587768555, "genuineness": 4.457139015197754, "identity": 0.48122209310531616, "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_AUTH_reg/7a0ef428ea7169c99eae0f28c987bd87c2863d703e34be52c6e87daca80687a1.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": "7a20494522f975c3b6d4577178332609d78301f158019c28dc3195256a897d5b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "899a78ad04ac320ca01ec579dd906431132a0ee712e49a4d4543cef0096787fc", "generation_seconds_batch": 24.988654232933186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.372091293334961, "genuineness": 3.9425034523010254, "identity": 0.6148620843887329, "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_AUTH_reg/7a20494522f975c3b6d4577178332609d78301f158019c28dc3195256a897d5b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "7a602f4d2d1036ae5aa927d80a269adc1b93a8dabffbbf6ca2230318ce70d747", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1975e31d5a6e1f2463b1f5b28fd39c3c637971703f0ba1320c834a7b06db1ee4", "generation_seconds_batch": 25.29584223101847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.443574905395508, "genuineness": 4.469690799713135, "identity": 0.5465295314788818, "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_AUTH_reg/7a602f4d2d1036ae5aa927d80a269adc1b93a8dabffbbf6ca2230318ce70d747.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "7b2e9d00deeca6cc211ec9abc36564a7339f10a501b180b257197e60fb39a575", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84ab087d2cde1b764b73fd16c22fd3a9d3a6e6b5562074b42bcb5f60e2074e3c", "generation_seconds_batch": 24.63029351597652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.149696350097656, "genuineness": 4.344170093536377, "identity": 0.7616534233093262, "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_AUTH_reg/7b2e9d00deeca6cc211ec9abc36564a7339f10a501b180b257197e60fb39a575.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "7b4bc49b18a0bda9bda08a16fd0fe591cf7147c2005d0629cef339615d3eec8a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5655658505af58ceb2b3a95f2d52d6618fca2c1e2d4a9da0d2c3df9d7740809a", "generation_seconds_batch": 25.01931671402417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.915895938873291, "target_z": 1.6406250197223484, "ce": 6.070269584655762, "genuineness": 4.068024158477783, "identity": 0.6795694828033447, "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_AUTH_reg/7b4bc49b18a0bda9bda08a16fd0fe591cf7147c2005d0629cef339615d3eec8a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "7b5c332db9fd7b372b88d3e2dbfd01f8d5bfe914b68e7f0beeaf68b6dce062b5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab25bd24ae9942de2dba5d1d2b13d18f07ecdefa19d2ce0893ea7d379f4e3881", "generation_seconds_batch": 25.590263979975134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5425496101379395, "target_z": 1.2812501541929053, "ce": 6.086154937744141, "genuineness": 4.18099308013916, "identity": 0.6101069450378418, "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_AUTH_reg/7b5c332db9fd7b372b88d3e2dbfd01f8d5bfe914b68e7f0beeaf68b6dce062b5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "7b651461e3214330cc9d5f2b83153b6fffeb3ffdb51976cc030ad7e91bf6876a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6006407feb3f6878affa5e12d39e7f8bcbd985d07c106138177192a078f2971f", "generation_seconds_batch": 26.843750506988727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 6.125868320465088, "genuineness": 3.9801597595214844, "identity": -0.060473036020994186, "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_AUTH_reg/7b651461e3214330cc9d5f2b83153b6fffeb3ffdb51976cc030ad7e91bf6876a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "7c96f9890faa5a879516f46ea9d9f243d4a44802e36412b81772dabd1beaa801", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "731f20b392fda184e4295d04a4a35032eaada1ac87713b4e9fadfffdebf87922", "generation_seconds_batch": 24.48547170497477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404573440551758, "target_z": 1.1484373646329726, "ce": 6.054384231567383, "genuineness": 4.557555198669434, "identity": 0.5650380849838257, "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_AUTH_reg/7c96f9890faa5a879516f46ea9d9f243d4a44802e36412b81772dabd1beaa801.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "7d3dd901afb68940ee6d2a426d1cfb2909ecc0c816990165159610f3b9847ef3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a80ca026631869a7ba74733538d81544b509483c5d6c5a022e643feafa7e8195", "generation_seconds_batch": 24.94242050102912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.510084629058838, "target_z": 1.2500000860611564, "ce": 6.332377910614014, "genuineness": 4.281409740447998, "identity": 0.4414992332458496, "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_AUTH_reg/7d3dd901afb68940ee6d2a426d1cfb2909ecc0c816990165159610f3b9847ef3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "7e44cb5e3e494586db037c947099a53fa0b73166f8fe0570d584b6947cc3008c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55c98a5e5dff1067fc1856026e5eeba04dc2294ebce66184003780df04ddf903", "generation_seconds_batch": 25.209855598048307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.038498878479004, "genuineness": 4.306513786315918, "identity": 0.5267075896263123, "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_AUTH_reg/7e44cb5e3e494586db037c947099a53fa0b73166f8fe0570d584b6947cc3008c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "7e4f654f769cd0d9f87a5ea132444d65b1510ee2fe5d4b4703f6be4a15ab0e64", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "527f6d517569f28a8ee0758721c01a8cb59edb5ee346467fa8e62a679e71f0b1", "generation_seconds_batch": 24.71703683992382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.284721851348877, "genuineness": 4.017816066741943, "identity": 0.18417972326278687, "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_AUTH_reg/7e4f654f769cd0d9f87a5ea132444d65b1510ee2fe5d4b4703f6be4a15ab0e64.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "7ee4acc50e6a0214cf1d15ef861b8445347238b42b4bfc78017d66ffe361c5d5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fadbe968d98942792d9049d86bfaf9f933956cd58b64bc97056fa0f614af99c9", "generation_seconds_batch": 25.52629535808228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.5368696451187134, "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_AUTH_reg/7ee4acc50e6a0214cf1d15ef861b8445347238b42b4bfc78017d66ffe361c5d5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "7ee7ba82cce3128e72ca77b20b194f70d3e3573d4b7150fb8499a28d8f0ec933", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06ce48bf6e11266fd38dad6792a805bf910b5224ddb192f5739fd8c3131f8a55", "generation_seconds_batch": 28.825104255927727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.078212261199951, "genuineness": 4.15588903427124, "identity": 0.4477956295013428, "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_AUTH_reg/7ee7ba82cce3128e72ca77b20b194f70d3e3573d4b7150fb8499a28d8f0ec933.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "7f07fa981e63800ce041ca9d11cc5c59186ef979287a292fc9040ee76199b1f7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8faa6bab887e23290c2f7a698992851ba22b1fc3ba53f26ab76e00cf6bfb58a2", "generation_seconds_batch": 26.691403631935827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.332377910614014, "genuineness": 3.9801597595214844, "identity": 0.2535678446292877, "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_AUTH_reg/7f07fa981e63800ce041ca9d11cc5c59186ef979287a292fc9040ee76199b1f7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "7f3922809417940eaab05989a8be88cf12744059c74db2cb4d2eff57bfd98640", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa1c37c837f2e61eb2380a7b08c1ed64dfce36119ca9565e4770e8158415440a", "generation_seconds_batch": 24.044044145033695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.38003396987915, "genuineness": 4.055472373962402, "identity": 0.2733445465564728, "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_AUTH_reg/7f3922809417940eaab05989a8be88cf12744059c74db2cb4d2eff57bfd98640.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "bcad2c0100559e6577fdd2afadde385d6e6c46deb58aad7a37989491f13d5061", "generation_seconds_batch": 16.221659452887252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9500648975372314, "target_z": 0.7109375582705747, "ce": 6.419747352600098, "genuineness": 4.425758361816406, "identity": 0.36761659383773804, "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_AUTH_reg/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "807538c24dd8100c1d17c9bc6d048dd8a653fedbd34f6febcf844158328e1e11", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3aefe71dcc1e10b2839a1f9f3844af27ea1b8f1faddb7132715b4fbc66361b2", "generation_seconds_batch": 26.138799433945678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.052167892456055, "target_z": 2.7343751093693864, "ce": 6.324435234069824, "genuineness": 4.218649387359619, "identity": 0.39702534675598145, "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_AUTH_reg/807538c24dd8100c1d17c9bc6d048dd8a653fedbd34f6febcf844158328e1e11.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "808ef82c0cfdd63f230d6c13c361e61864165319dd9a21a8575c27c6f1fa18c7", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 127, "seconds_generated": 10.16, "mp3_sha256": "22c919fd6c999376f6084a1ecb3481b7db1bd74b3511daa018a0c6f0362ef33f", "generation_seconds_batch": 18.507048763916828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.41180419921875, "genuineness": 4.049196243286133, "identity": 0.5263856649398804, "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_AUTH_reg/808ef82c0cfdd63f230d6c13c361e61864165319dd9a21a8575c27c6f1fa18c7.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "aa722f4b2444dbf98e978984c9f0c160f2f8b8614936a346c09e81a03a9142fc", "generation_seconds_batch": 15.72046716301702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7779200077056885, "target_z": 1.507812459658833, "ce": 6.348263263702393, "genuineness": 4.18099308013916, "identity": 0.5461978912353516, "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_AUTH_reg/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "810866b9ead2d0061695dbb91411c42708d9c7d46b2a0f0c1a3379ddd727e138", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4547a87eac9240816cc59b902e2b9e415aa55a07c9ed181cd646078a5e0aa687", "generation_seconds_batch": 28.825104255927727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.213237762451172, "genuineness": 4.124508857727051, "identity": 0.6609183549880981, "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_AUTH_reg/810866b9ead2d0061695dbb91411c42708d9c7d46b2a0f0c1a3379ddd727e138.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "818a74d8da51f87c3b2be2e8b52572a0186033195f55deec61f0c74ebc7b7b11", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e1d597e73f047ec3e57bfb48aa2a03935952377ae16d26aabbda8b117f6450d", "generation_seconds_batch": 27.106725885998458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0701041221618652, "target_z": 1.789062384355321, "ce": 6.0305562019348145, "genuineness": 4.369274139404297, "identity": 0.6218562126159668, "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_AUTH_reg/818a74d8da51f87c3b2be2e8b52572a0186033195f55deec61f0c74ebc7b7b11.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": "836c01d45d042304e125fff2a19eb519ec38a4660a7c8d26f14754bed99a4bb9", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41e5d09ada6f92aac62f234107dbc841b1103eb6bedb103fb7e15fb4a5f4fb33", "generation_seconds_batch": 16.86377843702212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.604071140289307, "target_z": 3.2656248906306136, "ce": 6.268836498260498, "genuineness": 4.093128204345703, "identity": 0.3968141973018646, "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_AUTH_reg/836c01d45d042304e125fff2a19eb519ec38a4660a7c8d26f14754bed99a4bb9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "8373b17e5bd0cfa907753fd6861943d04c67b6d0824a01ea39f720833b6f7bef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e2f49bcd603ab7fba26b25ee0bcfe00cf1609992d0a1e71f80486e568525e555", "generation_seconds_batch": 27.60471205005888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8428497314453125, "target_z": 1.5703123664259135, "ce": 6.09409761428833, "genuineness": 4.344170093536377, "identity": 0.7342202663421631, "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_AUTH_reg/8373b17e5bd0cfa907753fd6861943d04c67b6d0824a01ea39f720833b6f7bef.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "83ce3d3c22f535c2758e6677dc3a8448e66a7b4a008bd2334829a8687a4fa5a5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ce807211bdec431d85e9cd9e8230b9810e528fad64473c82a9d9486edd37a41", "generation_seconds_batch": 25.326602155924775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.292664527893066, "genuineness": 4.6203155517578125, "identity": 0.6574700474739075, "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_AUTH_reg/83ce3d3c22f535c2758e6677dc3a8448e66a7b4a008bd2334829a8687a4fa5a5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "845d7beeb97394fdeb469c56ac0afca7f26661c34b25add8c498cd93bd475e58", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8fb89f4fbb31d188ee641569d9e9e934746f72ca054d83e9c92af5b5228879b", "generation_seconds_batch": 25.43642353801988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.006728172302246, "genuineness": 4.344170093536377, "identity": 0.7433101534843445, "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_AUTH_reg/845d7beeb97394fdeb469c56ac0afca7f26661c34b25add8c498cd93bd475e58.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "8468bf9961e11a98e0f2c36684f4c690e9bb9d9bc40b3683cacd64d6eb5a0a06", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "215b2a84eb615410349312800ab8e3a90b95f85158c0ded156a4733b0fababde", "generation_seconds_batch": 27.38550260500051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.46386656165122986, "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_AUTH_reg/8468bf9961e11a98e0f2c36684f4c690e9bb9d9bc40b3683cacd64d6eb5a0a06.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "84da367a79a135359840a8f6063bde37b44e7f973b9fe75553bab4aa1e4de303", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4dea9d540e7b679830b5a14308bac192cbb5a59a9e57bdaaa29ba7ced5615a12", "generation_seconds_batch": 27.312456307001412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 5.903473377227783, "genuineness": 3.923675537109375, "identity": 0.7353646755218506, "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_AUTH_reg/84da367a79a135359840a8f6063bde37b44e7f973b9fe75553bab4aa1e4de303.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "8526f9f86dbbf582e81dcb698ec3a1a1509bc6da0f201b21e45afa5efabb9128", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 127, "seconds_generated": 10.16, "mp3_sha256": "994fe90b60ae48070d1a834d04cd598018148036f70fdce0ae561035cb764b3f", "generation_seconds_batch": 18.507048763916828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.316492557525635, "genuineness": 4.281409740447998, "identity": 0.3394375741481781, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschalten.", "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_AUTH_reg/8526f9f86dbbf582e81dcb698ec3a1a1509bc6da0f201b21e45afa5efabb9128.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "8535dc894648d50818f557b5fce711a0f733b9975e66b2e8e8f1d2b114695a74", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "4afecf23e6692d20999416c36eb7ec85d14fac020bec5f41bba60bcbf38b759c", "generation_seconds_batch": 20.067598817055114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.102569103240967, "target_z": 1.82031245248707, "ce": 6.078212261199951, "genuineness": 4.105680465698242, "identity": 0.6701900959014893, "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_AUTH_reg/8535dc894648d50818f557b5fce711a0f733b9975e66b2e8e8f1d2b114695a74.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "85ee50ac4cccff1671c12dc95312f9b15bd294217b47f84713db5f8528de4238", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8ce2f79545e1ba19050c45d10913292c175b5be8e32be6730ea910fbebc55429", "generation_seconds_batch": 26.69096199190244, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.100865364074707, "target_z": 2.7812502115670097, "ce": 6.181467056274414, "genuineness": 4.017816066741943, "identity": 0.2790520191192627, "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_AUTH_reg/85ee50ac4cccff1671c12dc95312f9b15bd294217b47f84713db5f8528de4238.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "8742155e7047b84000682aa052cfd68a3d0d0b5b4859bcb55c75c49ea5a63ecb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efe0a0cb2eb16e2b3f08c77b994d3e42afd62db20ead967b6dd01e0d4e95e519", "generation_seconds_batch": 25.731791501049884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 5.903473377227783, "genuineness": 3.9676074981689453, "identity": 0.76674485206604, "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_AUTH_reg/8742155e7047b84000682aa052cfd68a3d0d0b5b4859bcb55c75c49ea5a63ecb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "875014113a083b795dfec4be59a62faa4af2fadf561f1455ded440266d141e0c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39fbc0f94cae7ccca9c4ae205d66f4b9430a033fa64aa00be11a29a61648665f", "generation_seconds_batch": 25.255860126926564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4857358932495117, "target_z": 1.2265625349623448, "ce": 6.205295085906982, "genuineness": 4.570107460021973, "identity": 0.7506537437438965, "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_AUTH_reg/875014113a083b795dfec4be59a62faa4af2fadf561f1455ded440266d141e0c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "875c824e3b5f9d96c94c69458a090f3c81e9b0c660c0e0788218de13740a4814", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5186d256026da33fd7eda80ef7047190aded07b0a85439b4a47cf267b6c3851e", "generation_seconds_batch": 18.961955229984596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0701041221618652, "target_z": 1.789062384355321, "ce": 6.070269584655762, "genuineness": 4.381826400756836, "identity": 0.6747187376022339, "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_AUTH_reg/875c824e3b5f9d96c94c69458a090f3c81e9b0c660c0e0788218de13740a4814.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "87c5c13fd9b2a873d6e526dc7a51314c0e195bd12547a8324f35ed207b4f36be", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f773101a2df67553c6745f189c99a487cb822b393eb948b7926eb1ed80011f8", "generation_seconds_batch": 24.085979157942347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 6.2767791748046875, "genuineness": 4.2249250411987305, "identity": 0.3552314043045044, "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_AUTH_reg/87c5c13fd9b2a873d6e526dc7a51314c0e195bd12547a8324f35ed207b4f36be.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "8813a34cfdb99c8d581179823272533a9a5c883a5ee0e87152fe58136dbbbd95", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ea07f474176ae04e38ac70eb050a5c304c65bdd708d1574819c666ce19659b6", "generation_seconds_batch": 25.188401284976862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6967577934265137, "target_z": 1.429687518825878, "ce": 6.0305562019348145, "genuineness": 3.9801597595214844, "identity": 0.6428121328353882, "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_AUTH_reg/8813a34cfdb99c8d581179823272533a9a5c883a5ee0e87152fe58136dbbbd95.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "88437b218e81923ec71c2205c904c5a1dc0722e299d89b41dbd8522085a95810", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5267e60dd445483d9cadb725b91d0fa9b4a10358e6a6774654703e3d5a20361", "generation_seconds_batch": 25.48607203492429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.0305562019348145, "genuineness": 4.143336772918701, "identity": 0.6805092692375183, "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_AUTH_reg/88437b218e81923ec71c2205c904c5a1dc0722e299d89b41dbd8522085a95810.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "889a5fd3a880716a0557a7bc326f316d01f05507849e0ef9a494872baa138c3d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e73f8dc8c66372ac0d4644755bd381502a951da28a20da1aacc7d7df38c5b8d", "generation_seconds_batch": 26.081687061931007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.597659111022949, "target_z": 2.296875073510571, "ce": 5.919358730316162, "genuineness": 4.068024158477783, "identity": 0.6896389722824097, "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_AUTH_reg/889a5fd3a880716a0557a7bc326f316d01f05507849e0ef9a494872baa138c3d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "889ca5c0c80423af82c282f58da356223edbe80910253728aec51f9e6ec0ef97", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1842ffd3a4a99fd2d003d25ff7ac6b37e0261998c8a6f9d46cd95cf830705870", "generation_seconds_batch": 24.899244813947007, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8185012340545654, "target_z": 1.546875044823519, "ce": 6.173524379730225, "genuineness": 4.15588903427124, "identity": 0.701936662197113, "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_AUTH_reg/889ca5c0c80423af82c282f58da356223edbe80910253728aec51f9e6ec0ef97.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "89311b1120fc565593de16ff86b52acebed16aef7254493900fdd854e53176ff", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d3215af049b79b4c3101dd40d54894c4200cb970b410460fda4c7a01a72f7bd2", "generation_seconds_batch": 26.482058531953953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0538716316223145, "target_z": 1.7734373502894467, "ce": 6.268836498260498, "genuineness": 4.344170093536377, "identity": 0.5944005250930786, "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_AUTH_reg/89311b1120fc565593de16ff86b52acebed16aef7254493900fdd854e53176ff.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "8950758fa49dbc3fcb95030fa17d810ba1944e24ba1cad79f79bd6a75b7d046c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f33de6928e9dcdeb3e9d30b1c0d57fe9c275c5e5a8a3d1ea818ee583b2a9c1a7", "generation_seconds_batch": 19.916308910935186, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.891547203063965, "target_z": 1.6171874686235368, "ce": 6.181467056274414, "genuineness": 4.494794845581055, "identity": 0.4379402995109558, "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_AUTH_reg/8950758fa49dbc3fcb95030fa17d810ba1944e24ba1cad79f79bd6a75b7d046c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "896e49a56e9976890d1cd27bd01ec889d0398eb45b70fda1aec369032e2436f6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1da7b43ce1add0a8861eef6f298855c0f54bd123a1d96ecb452e6f852259a4bb", "generation_seconds_batch": 27.998524766997434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1188015937805176, "target_z": 1.8359374865529443, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.8404639959335327, "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_AUTH_reg/896e49a56e9976890d1cd27bd01ec889d0398eb45b70fda1aec369032e2436f6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "8a9df7b40444e92a049d91ccfb07919459e46830a4ff636a643998a2cbbe70ce", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "522b7f9fc340f4f3088854a3ebb37cd3283a916b386b7baabe136eabd66323a6", "generation_seconds_batch": 25.43642353801988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.086336612701416, "target_z": 1.8046874184211954, "ce": 6.09409761428833, "genuineness": 4.21237325668335, "identity": 0.6104329824447632, "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_AUTH_reg/8a9df7b40444e92a049d91ccfb07919459e46830a4ff636a643998a2cbbe70ce.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "8b8cd3128ac0597f6f5d7865c608d76152576b81240b74fc24ee949ee4107ff9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "82288ebc2988f29fc87e517a73bbb7fc14779bc1fcaf1210dd3ce107cb18cb7c", "generation_seconds_batch": 25.422244618996046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.5689942836761475, "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_AUTH_reg/8b8cd3128ac0597f6f5d7865c608d76152576b81240b74fc24ee949ee4107ff9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "8bbdc5ceeb7680c67d4679ce658d296f4004e977e978544c7379c0e31bd0590c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a106ab8f3de956aec360392d5579ca0627b7a4a3139cc23d190dd3ca77b78f20", "generation_seconds_batch": 25.790932217030786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.435632705688477, "genuineness": 3.936227321624756, "identity": 0.5589097738265991, "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_AUTH_reg/8bbdc5ceeb7680c67d4679ce658d296f4004e977e978544c7379c0e31bd0590c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "8bcb9150edad79aff29a6a0a7155ff53cabf204bde0c5eeb8c9a412bc149b28a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d0d433bfe1abdeaf7ccf916fabb670a7b6d10803fbd390dc10a803c41cf326bc", "generation_seconds_batch": 25.743776886025444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 6.181467056274414, "genuineness": 3.9550557136535645, "identity": 0.09842225909233093, "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_AUTH_reg/8bcb9150edad79aff29a6a0a7155ff53cabf204bde0c5eeb8c9a412bc149b28a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "8beb8cadb1e2978ce8677001e11b09624324d48cc3dbf7b7ab6f79b4807970ba", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9da806b04c18bfeaa007bbde771875a3b567604d7bd4d6446e5e8ce3faadccd4", "generation_seconds_batch": 28.117583174956962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5425496101379395, "target_z": 1.2812501541929053, "ce": 6.427689552307129, "genuineness": 4.406930446624756, "identity": 0.29634958505630493, "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_AUTH_reg/8beb8cadb1e2978ce8677001e11b09624324d48cc3dbf7b7ab6f79b4807970ba.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "8c0bdba73a3d711ac62424e8fc39dc4754a0652c49d32fcfb04a313b58def5ef", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "764d89019eef1e039a12ec62a57a6ffd0ca59a2b5e783f7c38a574eee516aa21", "generation_seconds_batch": 25.923318175948225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0132906436920166, "target_z": 1.7343749946211777, "ce": 6.260893821716309, "genuineness": 4.2249250411987305, "identity": 0.1704835444688797, "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_AUTH_reg/8c0bdba73a3d711ac62424e8fc39dc4754a0652c49d32fcfb04a313b58def5ef.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "8c24864081726fe12e5d522b5198ff6063bf7c799e4fad78dfd42c5973a03106", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64338428a2d12e2917ed9e8dd7fbc56ad081dd205f67e2edbc21e2aa761a3c7e", "generation_seconds_batch": 24.25232342001982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.41180419921875, "genuineness": 4.20609712600708, "identity": 0.6215404272079468, "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_AUTH_reg/8c24864081726fe12e5d522b5198ff6063bf7c799e4fad78dfd42c5973a03106.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "8ca082c25039115b8fbd431d9e92e340a276925a279077ddd0ffe6f66761f30c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19a98e2831d811660180f97d27ad558e99c36197886f9dd2104e54d343125b9f", "generation_seconds_batch": 25.477295704069547, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.493852138519287, "target_z": 1.234375051995282, "ce": 6.133810997009277, "genuineness": 4.319066047668457, "identity": 0.7354081869125366, "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_AUTH_reg/8ca082c25039115b8fbd431d9e92e340a276925a279077ddd0ffe6f66761f30c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "8d0b96d5f3238476099d9c4e652d43fc5b09fe8f8ca6fd561c1a2791374aa0aa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d0982d97bf55e4dd3b57db618eb55347010b1039faf36cfcfe5cc2f986ca28c", "generation_seconds_batch": 27.05371477000881, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.607978105545044, "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_AUTH_reg/8d0b96d5f3238476099d9c4e652d43fc5b09fe8f8ca6fd561c1a2791374aa0aa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "8da146637a22b6caab516d98ed06f294facf4065a1d85311daa36912b8e911b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbd2a8ab2a05f703bbe23db6f8f261dc07e858d84b770f241cc426ba61755dfc", "generation_seconds_batch": 18.633984974003397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.021406888961792, "target_z": 1.742187511654115, "ce": 6.284721851348877, "genuineness": 4.406930446624756, "identity": 0.4255982041358948, "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_AUTH_reg/8da146637a22b6caab516d98ed06f294facf4065a1d85311daa36912b8e911b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "8e88219df71e58277704633f37669c0387ffdf6b0351451980c81b77204592c7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64017b5f02e9641b02709c9e2ee3c915ce886c8ef0d8accc704916b31ff2039f", "generation_seconds_batch": 26.515055862953886, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.038498878479004, "genuineness": 4.143336772918701, "identity": 0.22764775156974792, "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_AUTH_reg/8e88219df71e58277704633f37669c0387ffdf6b0351451980c81b77204592c7.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": "8efa5e63ae59d959b4b8c8be51a13df2f89f29d1ee821063cdc87ce43a85afd3", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6c481c8f121c17ab91b1f892fd795bbf170911d7aa381e4fbc5c62e2e3d89356", "generation_seconds_batch": 16.378719346947037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.2767791748046875, "genuineness": 4.105680465698242, "identity": 0.7271194458007812, "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": 0.03999999999999915, "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_AUTH_reg/8efa5e63ae59d959b4b8c8be51a13df2f89f29d1ee821063cdc87ce43a85afd3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "8f268188263a0de8f8b190e4c36b4adf9e0483b7cf4fff501e73577afc71d28f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c47e607311492a917db005abeb6232fb57b8ed9c293252d48b01e592f97e68a4", "generation_seconds_batch": 21.847837987937964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.046441555023193, "genuineness": 4.268857479095459, "identity": 0.5806360244750977, "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_AUTH_reg/8f268188263a0de8f8b190e4c36b4adf9e0483b7cf4fff501e73577afc71d28f.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": "8f766145f155c1e442b9d9e94dfbe12f3a23e213dc0870a660e6c751d8ce75c8", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "16cc5c20d416ba06d7272c6f487de7296a804b892e988f87d624ea57c3d6a3b8", "generation_seconds_batch": 24.24600269505754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.052167892456055, "target_z": 2.7343751093693864, "ce": 6.403861999511719, "genuineness": 4.105680465698242, "identity": 0.43264710903167725, "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_AUTH_reg/8f766145f155c1e442b9d9e94dfbe12f3a23e213dc0870a660e6c751d8ce75c8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "8f890bb22d9877d9a3133c37bd7116274362337b2ce51905f19eba9e174c9dc6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3bd253af85455690726bb5323eb9753a615724915d93b48ae7cd2294da2275c", "generation_seconds_batch": 25.023820757051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.1020402908325195, "genuineness": 4.293961524963379, "identity": 0.6393859386444092, "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_AUTH_reg/8f890bb22d9877d9a3133c37bd7116274362337b2ce51905f19eba9e174c9dc6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "900ecf497779f6aa3c576ffe8a8140bbaf4a751d206eb256f23e6e36fdc6bc1c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bc9fdb7083c5d8c5ce9f77e63822b71b9f343e40e83fcc104a06492bd3920f28", "generation_seconds_batch": 20.302960409899242, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.575014352798462, "target_z": 1.312499992828237, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.7257845401763916, "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_AUTH_reg/900ecf497779f6aa3c576ffe8a8140bbaf4a751d206eb256f23e6e36fdc6bc1c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "90724beb4b8ef20b5ca7b1eb784affc4587aff567cad70134ff813ccaba66286", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1683bd9f20a0f3baef78940e317db99c3fc69bf99abe6c11a0a4f5e56e1d21b", "generation_seconds_batch": 19.99374872399494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.181467056274414, "genuineness": 4.306513786315918, "identity": 0.7753146886825562, "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_AUTH_reg/90724beb4b8ef20b5ca7b1eb784affc4587aff567cad70134ff813ccaba66286.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "90fb4d9063874911fb21810b1b7fabeaaf42a0233c7359c8ccb2bd7176da5942", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02847a252daa54c40ce0a25734283524a40e0dacae0014e47ce5f8f646765cd6", "generation_seconds_batch": 24.732398508000188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.348263263702393, "genuineness": 4.419482707977295, "identity": 0.7185330390930176, "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_AUTH_reg/90fb4d9063874911fb21810b1b7fabeaaf42a0233c7359c8ccb2bd7176da5942.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "920f4e54bde999caf9e9348352a443dbfaf7b93d533fc5cbc49c0cbb2123a59b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86ba776651cb96a4b939f083bb12a2a5be5d983c814522840aee1ae6c148d7ec", "generation_seconds_batch": 25.0076725439867, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.173524379730225, "genuineness": 4.1998209953308105, "identity": 0.45667529106140137, "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_AUTH_reg/920f4e54bde999caf9e9348352a443dbfaf7b93d533fc5cbc49c0cbb2123a59b.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": "935dabbb0d33f499bc5f2d79c981fdb0c2653f69f9a99419a3cb75c1c8062b48", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61812399e3122fbb9c5d826f1f84769b957b758cad148a02374cd0541fce3912", "generation_seconds_batch": 15.495892810984515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.889843463897705, "target_z": 2.5781252277034765, "ce": 6.324435234069824, "genuineness": 3.8922953605651855, "identity": 0.3284645676612854, "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_AUTH_reg/935dabbb0d33f499bc5f2d79c981fdb0c2653f69f9a99419a3cb75c1c8062b48.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": "9370ab04fb04601c6f08c0b897c43a06ecf8142c2a68adb0aab7a60513fcf80b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dadcd9c2317aeced76e0291852f8fa302e14c18f338457bc2bf28ea804827add", "generation_seconds_batch": 19.763231093995273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8590822219848633, "target_z": 1.5859374004917879, "ce": 6.443574905395508, "genuineness": 4.18099308013916, "identity": 0.5498694777488708, "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_AUTH_reg/9370ab04fb04601c6f08c0b897c43a06ecf8142c2a68adb0aab7a60513fcf80b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "9399c232f3408fffe624df79726c041d17e778d64c1df65f30451ade4d3c9a31", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e248f475643d41da0d6f91be30000b525abf8e28908d403559199ada9658922e", "generation_seconds_batch": 25.382786588044837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.109982967376709, "genuineness": 4.042920112609863, "identity": 0.8382853865623474, "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_AUTH_reg/9399c232f3408fffe624df79726c041d17e778d64c1df65f30451ade4d3c9a31.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "93e465a7bda2c01c5782ff4b867217e92d934a985b6b92e13cae8c2d5992a48a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b091458c556689e85a339bfcb7d34b39cbd3abcb477e8485ab069aceb37ca11e", "generation_seconds_batch": 18.961955229984596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932128429412842, "target_z": 1.6562500537882228, "ce": 6.292664527893066, "genuineness": 4.482243061065674, "identity": 0.6161745190620422, "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_AUTH_reg/93e465a7bda2c01c5782ff4b867217e92d934a985b6b92e13cae8c2d5992a48a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "941eb1617b99db6f4506f0d3fc34b02c8e9196cdbfd0b26f6ab036486e3a2365", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acbd5542de1c8c3daa41eaf9f34392b89538034e890d5ac638ba1737f665f43a", "generation_seconds_batch": 25.296295181964524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.425514221191406, "target_z": 3.0937499748988295, "ce": 6.1020402908325195, "genuineness": 4.030367851257324, "identity": 0.5245511531829834, "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_AUTH_reg/941eb1617b99db6f4506f0d3fc34b02c8e9196cdbfd0b26f6ab036486e3a2365.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "94498295aac5205ce6fcb6b3108c620f26bfb51d14b14ef51660a22ca1b943f7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b96a6cf2ac902aa756127808c0f48af7dd8dd4612e64f3c99a3ae10bcb0cd680", "generation_seconds_batch": 24.241931298049167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.173524379730225, "genuineness": 4.017816066741943, "identity": 0.17293623089790344, "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_AUTH_reg/94498295aac5205ce6fcb6b3108c620f26bfb51d14b14ef51660a22ca1b943f7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "94f190dada4fc2845ff96b1751468f708a58a28134c5418288866ded1e569748", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f09e8738d58100f1758300b1fb89259e510023df82a571b5a80fa9042b058f46", "generation_seconds_batch": 18.95389297895599, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.221180438995361, "genuineness": 4.381826400756836, "identity": 0.19692906737327576, "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_AUTH_reg/94f190dada4fc2845ff96b1751468f708a58a28134c5418288866ded1e569748.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": "950ffb8ead53ad3f7bad573703d53a888957f2a645e490be95895f2b0660621a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "567be3e6b082179d3b123bc1411d5146bd157278d700a8602e9e6b57b21b40ee", "generation_seconds_batch": 16.86377843702212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393049240112305, "target_z": 3.0624999067670804, "ce": 6.24500846862793, "genuineness": 4.036643981933594, "identity": 0.4191948175430298, "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_AUTH_reg/950ffb8ead53ad3f7bad573703d53a888957f2a645e490be95895f2b0660621a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "954b140acb07664b52a03709164df6a9ba5395d15c073314db705539296a972c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fae0c5a29d915b97dd07f8f38c5c162a80137841672383a04436eb8f0532dad8", "generation_seconds_batch": 27.393418796011247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 5.935244083404541, "genuineness": 3.9550557136535645, "identity": 0.7824923992156982, "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_AUTH_reg/954b140acb07664b52a03709164df6a9ba5395d15c073314db705539296a972c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "9566431fb44071ecc16dd25c6b3f42a9359217c8098663b134838a1247bfd016", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "524ca15323d3a53755adc33060e5f50b2d9ebf270fdd37bc3eab691d4e613344", "generation_seconds_batch": 26.108274420024827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.071808338165283, "target_z": 0.8281250842682156, "ce": 6.348263263702393, "genuineness": 4.432034492492676, "identity": 0.5235214233398438, "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_AUTH_reg/9566431fb44071ecc16dd25c6b3f42a9359217c8098663b134838a1247bfd016.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": "9568edef3c756f16a89b3a6e82f5fd6e7d8a0138940d75ab3f952c6c65dd4a6a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87d32001cb2dc4942983c87b9ef8c2cd07320d60a22ddaf07976ade5843a16da", "generation_seconds_batch": 17.19196807104163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950535774230957, "target_z": 2.3906248189129835, "ce": 6.316492557525635, "genuineness": 3.8671910762786865, "identity": 0.5425817370414734, "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_AUTH_reg/9568edef3c756f16a89b3a6e82f5fd6e7d8a0138940d75ab3f952c6c65dd4a6a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "96741628dda7d2aec91bf08e4feb5cfdbf1df14e3a1ebcc43b21e1679d25eed9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68b707a67d23c079bef6b7d37cb30be6f928c68b2db8dea5a4b3a4c102a70797", "generation_seconds_batch": 25.49590427801013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.980825901031494, "target_z": 1.703125155985846, "ce": 6.284721851348877, "genuineness": 4.337893962860107, "identity": 0.7923809289932251, "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_AUTH_reg/96741628dda7d2aec91bf08e4feb5cfdbf1df14e3a1ebcc43b21e1679d25eed9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "96eab3c4bcf3bf18f940806e759739c423935abdf4bf49d6fed95077f2880269", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41846a4a5777bc8ea5274c07db3cbeec6c776103bf27d619690b3557d9d9889c", "generation_seconds_batch": 25.358860118081793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.197352409362793, "genuineness": 4.268857479095459, "identity": 0.6022040843963623, "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_AUTH_reg/96eab3c4bcf3bf18f940806e759739c423935abdf4bf49d6fed95077f2880269.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "97a6d67248818605348866bdd73fbbb11cd13713d27e97192932621ebe05b5ca", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03be8accfa5b45793c12d31108c64f4e05084261642c04cff937468cb4a30270", "generation_seconds_batch": 25.83499683998525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.332377910614014, "genuineness": 4.080576419830322, "identity": 0.6068015694618225, "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_AUTH_reg/97a6d67248818605348866bdd73fbbb11cd13713d27e97192932621ebe05b5ca.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "97ca11d04de025fda5724c1e9ed115800154939643a3f641ad9c42ccef79e61a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2959170e01fea5217e9104921600b516cec740184518ba6f09d4c846cd0be0e", "generation_seconds_batch": 25.49590427801013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149562358856201, "target_z": 2.8281248547717985, "ce": 6.165581703186035, "genuineness": 4.18099308013916, "identity": 0.47373223304748535, "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_AUTH_reg/97ca11d04de025fda5724c1e9ed115800154939643a3f641ad9c42ccef79e61a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "97f664cf3ded8eb4e1c060c71bffbc7c171b3163df81562c8577452d8af349e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15770a28302efd0a1dc48495d0887e87d3de716f1420db3a0ba30dd8574697b3", "generation_seconds_batch": 26.054093715036288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.221180438995361, "genuineness": 3.810706615447998, "identity": 0.18125012516975403, "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_AUTH_reg/97f664cf3ded8eb4e1c060c71bffbc7c171b3163df81562c8577452d8af349e5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "97fa33daf0dac23f9575d3ad03e7bf82883abdd2040e0bfa6ce8e37f7c5da25d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d19ffd49463c553bdf02a5fea7c2a7b2e3c6577f03e815f74c3410294643110a", "generation_seconds_batch": 24.734309680992737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.070269584655762, "genuineness": 4.400654315948486, "identity": 0.6507788300514221, "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_AUTH_reg/97fa33daf0dac23f9575d3ad03e7bf82883abdd2040e0bfa6ce8e37f7c5da25d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "989693bb354d7ffde4ca3d23cf48105e1f3d99af2359233e25961eab17e9c0f9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0da809632cbb677b6ad4d56e2f4bf9673461092f4d31a55a11e849f2b8c255a0", "generation_seconds_batch": 20.183363327058032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.308549880981445, "genuineness": 4.118232727050781, "identity": 0.522799551486969, "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_AUTH_reg/989693bb354d7ffde4ca3d23cf48105e1f3d99af2359233e25961eab17e9c0f9.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": "98a51f074b3117c4264df66c5c35333cecef414ef225dab21ab0cc83286c366a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73f5a3e463192f9d8eb12eba3cb6095d7adba3af301161daf35bc67bb16abaea", "generation_seconds_batch": 16.469661337905563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9077796936035156, "target_z": 1.6328125026894111, "ce": 6.38003396987915, "genuineness": 4.15588903427124, "identity": 0.7546821236610413, "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_AUTH_reg/98a51f074b3117c4264df66c5c35333cecef414ef225dab21ab0cc83286c366a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "992c67edac369e64bf61855afd885cbb6ab37aa62a137eec226fd21fda3af9bf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "954292b34242045eaf461cfd591ad5ce4e8f25ea2a6e649d8115791db95232a6", "generation_seconds_batch": 26.15257835597731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5019683837890625, "target_z": 1.2421875690282191, "ce": 5.577824115753174, "genuineness": 4.243753433227539, "identity": 0.778927206993103, "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_AUTH_reg/992c67edac369e64bf61855afd885cbb6ab37aa62a137eec226fd21fda3af9bf.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": "99544f16ffba2474729d7a2684cf8bf035374ad6cb5f52a49695aa91b4e25035", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d64b6d6d277f56896b593c2671700db1faa9f3b42855ea37ac8f54a69b5f3944", "generation_seconds_batch": 16.5093660390703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.0305562019348145, "genuineness": 4.105680465698242, "identity": 0.46362388134002686, "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_AUTH_reg/99544f16ffba2474729d7a2684cf8bf035374ad6cb5f52a49695aa91b4e25035.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "cf7e5b54a1b57d740a11d48465b6ad4d45344038846698933e33706ec3888e8e", "generation_seconds_batch": 16.30944906908553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1164474487304688, "target_z": 0.8710936984529531, "ce": 6.324435234069824, "genuineness": 4.7458367347717285, "identity": 0.6590344905853271, "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_AUTH_reg/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "99f59c8db6b32213538504fe18edb7953b8af0753942c80b8805362c62f15a5d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e48ff32fb30fbb2341b1f75c160459cdc797a0f8c71b47c7e96fb2c4a94beaa", "generation_seconds_batch": 26.094968223012984, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.054384231567383, "genuineness": 4.331617832183838, "identity": 0.6919184327125549, "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_AUTH_reg/99f59c8db6b32213538504fe18edb7953b8af0753942c80b8805362c62f15a5d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "9a3e40d57bbaab39b27a45f1f80a367f3d72e5badce89bbcf26e95467906274f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28fcdb1386ad524c4c471e4346312d19cc2c10d4dbcf36a3e0353d03d559992d", "generation_seconds_batch": 25.023820757051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9240121841430664, "target_z": 1.6484375367552855, "ce": 6.149696350097656, "genuineness": 4.344170093536377, "identity": 0.40608981251716614, "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_AUTH_reg/9a3e40d57bbaab39b27a45f1f80a367f3d72e5badce89bbcf26e95467906274f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "9a4490ab4c2b97c1cccb4922f2be0240138d6344b2945f8b469cb463636d7a37", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2b719e92373652b068b3ba284bcb62c507aaabc9cf6b64210f39921c1daade4a", "generation_seconds_batch": 24.332862335955724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.125868320465088, "genuineness": 4.193545341491699, "identity": 0.5823668241500854, "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_AUTH_reg/9a4490ab4c2b97c1cccb4922f2be0240138d6344b2945f8b469cb463636d7a37.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "9aef69164d592662696ca19785d11159bfde02154e5cd30575684503f1415bfc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05678c425968dcefeba32c1219b4d66240ee9622e1bde38582e302183cd9be76", "generation_seconds_batch": 25.519448391045444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 6.117925643920898, "genuineness": 4.2751336097717285, "identity": 0.774007260799408, "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_AUTH_reg/9aef69164d592662696ca19785d11159bfde02154e5cd30575684503f1415bfc.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": "9b3c8ac304232639f616202e14ce17d33dccb123e0e019f01ba796162986cfd4", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "766f51c83700c6e3737aeee24fb6877d2ca9af7048451845407aac5536c481ff", "generation_seconds_batch": 16.516120343003422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.7574388980865479, "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_AUTH_reg/9b3c8ac304232639f616202e14ce17d33dccb123e0e019f01ba796162986cfd4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "9b44a6e87a71bfe828a3663914e742e02ccce92e5e8d5605e90c71dbf2a9e913", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c8648ee4ac1e9a0806912647b999122691b8ec71fa0629d7d00633ea74ba30f", "generation_seconds_batch": 27.106725885998458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.046441555023193, "genuineness": 4.105680465698242, "identity": 0.6282526254653931, "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_AUTH_reg/9b44a6e87a71bfe828a3663914e742e02ccce92e5e8d5605e90c71dbf2a9e913.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "9b966be16b8aea6c4dd0f614d4f29112fc1036d7c5ad969519aa2d29ef72d5a7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cebf5a6cd3bf1758c36db2df0e4ca5d52b5edb0996bc3c74427bfc8dc9f29fa", "generation_seconds_batch": 20.705880695022643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.284721851348877, "genuineness": 4.444586753845215, "identity": 0.7954943180084229, "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_AUTH_reg/9b966be16b8aea6c4dd0f614d4f29112fc1036d7c5ad969519aa2d29ef72d5a7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "9b9e5e4dce897aefa3cfe70e5dcdf4ba681e791e126e79da98dc89c1e5d2932c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "391b64d9d0f02012356fde4e85c2fd49c412bc7877c8b25bc816efaa48b66ad3", "generation_seconds_batch": 27.01790074305609, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 5.895530700683594, "genuineness": 3.9927117824554443, "identity": 0.6425615549087524, "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_AUTH_reg/9b9e5e4dce897aefa3cfe70e5dcdf4ba681e791e126e79da98dc89c1e5d2932c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "9c06fa9c41c68eeb34e9e2c2d1f6548a406697868b868179a31a4bee8533916b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "087e0c7be526ca0b07545d4bf3af980ebdf6f446a52f7219844fad7d927b3731", "generation_seconds_batch": 26.98474163992796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8022687435150146, "target_z": 1.5312500107576446, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.6568253040313721, "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_AUTH_reg/9c06fa9c41c68eeb34e9e2c2d1f6548a406697868b868179a31a4bee8533916b.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "5813213ad5b0b0e6ef23660e37a11568ca31ff342339173f17480806dde69824", "generation_seconds_batch": 19.145839125965722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.5596579909324646, "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_AUTH_reg/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "9c3388856e52330f993be11fdd50b9d721702405a26f56b6ecd8b0b18451014b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a870f684140301b70d44337be928a4e633290db1a56c5e636e122f5575ef38e7", "generation_seconds_batch": 18.83484937006142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.412689685821533, "target_z": 1.15624988166591, "ce": 6.2767791748046875, "genuineness": 4.381826400756836, "identity": 0.2639593183994293, "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_AUTH_reg/9c3388856e52330f993be11fdd50b9d721702405a26f56b6ecd8b0b18451014b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "9c8beae07b17824fd4b24edde1bd1e3cb1375bbbba981a27ed974ab3d4e6837e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70174238d800ebd4133d201602b21d9007095c157b5e01c3e92287abaf26f72a", "generation_seconds_batch": 24.94242050102912, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6318278312683105, "target_z": 1.3671873825623804, "ce": 6.348263263702393, "genuineness": 4.18099308013916, "identity": 0.43320804834365845, "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_AUTH_reg/9c8beae07b17824fd4b24edde1bd1e3cb1375bbbba981a27ed974ab3d4e6837e.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "e723e32c3f6f7496f2f7470c21db93b1eececd61479ed9a6c452aab570a0883a", "generation_seconds_batch": 18.08032832108438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.639944076538086, "target_z": 1.3749998995953174, "ce": 5.982900142669678, "genuineness": 4.293961524963379, "identity": 0.631820797920227, "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_AUTH_reg/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.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": "9cdb3f5ed8c8b0f592dbb88ec50e000ebe5f5b8d03d37fd246a8e816e7f6045c", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dadba1b9356b2d04140b6821704f77659e2dadf5b68dc27c4c19edfc03b3e287", "generation_seconds_batch": 22.072075044037774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.0305562019348145, "genuineness": 3.923675537109375, "identity": 0.7611967921257019, "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_AUTH_reg/9cdb3f5ed8c8b0f592dbb88ec50e000ebe5f5b8d03d37fd246a8e816e7f6045c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "9ce023f838b4efc285d070d0d6097700d44cc70abcb94aa12c9282eaf30f4e61", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2659524ed0ff644db55b51ba466f40e495f2fb815e6f2e13177f033716db4b55", "generation_seconds_batch": 20.82471571408678, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.181467056274414, "genuineness": 4.287685871124268, "identity": 0.5547454357147217, "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_AUTH_reg/9ce023f838b4efc285d070d0d6097700d44cc70abcb94aa12c9282eaf30f4e61.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "9d84dc5db2fea081e333928e69fb10381b3a1ef54282d9c014aa6737fc9f10be", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "b435f22673fff4188ff5b26d77cd6f5d17e3c0da86b1881e7e4ac07b6f7f3f1e", "generation_seconds_batch": 19.060482084052637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0312271118164062, "target_z": 0.7890624991035297, "ce": 6.316492557525635, "genuineness": 4.457139015197754, "identity": 0.14980503916740417, "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_AUTH_reg/9d84dc5db2fea081e333928e69fb10381b3a1ef54282d9c014aa6737fc9f10be.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "9d9f2b4a9c9ec14e9a86b5736ef3def650ea7a04db561a02866bcfec985df86e", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "e004b9b39a7f547837ee98fbe6e3993b424e6ebae9b80e193ae17b9e3fb6aabe", "generation_seconds_batch": 19.916043839999475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.599363088607788, "target_z": 1.3359375439270487, "ce": 6.078212261199951, "genuineness": 4.306513786315918, "identity": 0.7212806940078735, "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_AUTH_reg/9d9f2b4a9c9ec14e9a86b5736ef3def650ea7a04db561a02866bcfec985df86e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "9e5b1cb11fb0cc90a6e9ed7dafb8f9eb1e1e3d41d451894db2538d2a326c8019", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "160635413eeae1baabb22ddbcb3eb61739c37c9dd34203a60d31bb3e19830450", "generation_seconds_batch": 26.036088299006224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2405447959899902, "target_z": 1.9531247830541683, "ce": 6.141753673553467, "genuineness": 3.8922953605651855, "identity": 0.32696282863616943, "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_AUTH_reg/9e5b1cb11fb0cc90a6e9ed7dafb8f9eb1e1e3d41d451894db2538d2a326c8019.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "9e5e97978b7bac8cd43abb3a11e62d324ce811ffed7ac5e57a3f4d55268ebd4e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35656d8485b59c1ac35fcc4fdf631193a199dd43c59086bd360a671bf82e6c92", "generation_seconds_batch": 24.53017605491914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.117925643920898, "genuineness": 4.356721878051758, "identity": 0.5886374115943909, "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_AUTH_reg/9e5e97978b7bac8cd43abb3a11e62d324ce811ffed7ac5e57a3f4d55268ebd4e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "9e92ebc5f41e135b8775f02e0b59004c74b3d7d717d74155f14a515deafef9b1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "14cae183fc8f5a1b857ad87cbea0d88e902a11eed4410291d47982cfd46e5c75", "generation_seconds_batch": 17.954380632028915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.234132766723633, "target_z": 0.9843749659341255, "ce": 6.149696350097656, "genuineness": 4.494794845581055, "identity": 0.7725027799606323, "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_AUTH_reg/9e92ebc5f41e135b8775f02e0b59004c74b3d7d717d74155f14a515deafef9b1.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": "9e94e9ae0deea2de1dbaaf063b6749dda2865b6b0473b61c6fa34d6cce232264", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee3ac348e7449f2d1740f0162a963fac20eb75824410ce12193c991519eb2acc", "generation_seconds_batch": 15.495892810984515, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.403861999511719, "genuineness": 3.8671910762786865, "identity": 0.4210869073867798, "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_AUTH_reg/9e94e9ae0deea2de1dbaaf063b6749dda2865b6b0473b61c6fa34d6cce232264.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "9edc1a0c929c5b109d289f5852bf79c4c8e642735dc9028b225b4a8f3fd23bae", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "957be7c6378f89f875f65fe4d295993a0b3ed6912a3f6aa5dd92ed8ab5ad74ff", "generation_seconds_batch": 25.407685216050595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.054384231567383, "genuineness": 4.218649387359619, "identity": 0.6164780855178833, "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_AUTH_reg/9edc1a0c929c5b109d289f5852bf79c4c8e642735dc9028b225b4a8f3fd23bae.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "9f32efb1582e825c07188b0a05fe33a0c657bf43204618701ce132d97fcf679c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "498d43506a1db19ad6cfab3c9f91f4779c7d20eaae8ea7952b0d95937ed6852c", "generation_seconds_batch": 25.669120797072537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8590822219848633, "target_z": 1.5859374004917879, "ce": 6.3641486167907715, "genuineness": 4.231201171875, "identity": 0.3036339581012726, "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_AUTH_reg/9f32efb1582e825c07188b0a05fe33a0c657bf43204618701ce132d97fcf679c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "9f45398a58a7184a1aa831527cf2d7b0e78fdda68f4b3312d8c336e6a88a9043", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d88b8413aa17f5ec0cee7488df8bcd14a1e82f2a13ea821e3968ccad2ac7454b", "generation_seconds_batch": 30.237586554023437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.252951145172119, "genuineness": 4.13078498840332, "identity": 0.16087234020233154, "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_AUTH_reg/9f45398a58a7184a1aa831527cf2d7b0e78fdda68f4b3312d8c336e6a88a9043.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": "9f8f8dc8c83a608afc6729763b4fcb2e785ed2fbb297fc728961b970cbdb5cce", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e358d5055b9b3d439b01e95a0a6464b14772f58b9876b180a69321d7b3643dcf", "generation_seconds_batch": 16.48313725506887, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.316492557525635, "genuineness": 4.20609712600708, "identity": 0.7784066796302795, "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_AUTH_reg/9f8f8dc8c83a608afc6729763b4fcb2e785ed2fbb297fc728961b970cbdb5cce.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "a0f3621b736bde018195207ca173e59932d1e65efb8f1c1a789b21acb372bcfd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0a58cad3cd632d40d8cd05abb30093f87a88a0f851a8fc837ecd37f79f02063", "generation_seconds_batch": 25.365654174005613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9240121841430664, "target_z": 1.6484375367552855, "ce": 6.41180419921875, "genuineness": 4.162164688110352, "identity": 0.2722353935241699, "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_AUTH_reg/a0f3621b736bde018195207ca173e59932d1e65efb8f1c1a789b21acb372bcfd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "a114db42adc14fda413cbde7f30c9cd93f4affd73c7883d3e6236564a4e9995d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "223b2fcd85b5ac945a764fa1a8ca7c345554b10ce40a7bf7f99fe23f50d4b058", "generation_seconds_batch": 26.373753060004674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.0305562019348145, "genuineness": 4.231201171875, "identity": 0.7289023995399475, "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_AUTH_reg/a114db42adc14fda413cbde7f30c9cd93f4affd73c7883d3e6236564a4e9995d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "a133b2f3952862441a4ffd059a0cf589faa6582de8383c376dc73350ae4d48dd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "26c67ec53c872b8612e70438bbb702a7fa30e0ccd624dba877546dc57540b55f", "generation_seconds_batch": 28.278084215940908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402869701385498, "target_z": 2.1093751237129124, "ce": 6.165581703186035, "genuineness": 3.7165660858154297, "identity": 0.5354949235916138, "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_AUTH_reg/a133b2f3952862441a4ffd059a0cf589faa6582de8383c376dc73350ae4d48dd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "a1579d23acec6ada6e7c8701f13d1ae431a94a74d68981a63cc081fb06f1c80d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cc26f870ad0978d3de9125d1e698970bb60b22db8994ead4aaec41d30da5f60", "generation_seconds_batch": 25.257621766999364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.213237762451172, "genuineness": 3.7165660858154297, "identity": 0.25134149193763733, "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_AUTH_reg/a1579d23acec6ada6e7c8701f13d1ae431a94a74d68981a63cc081fb06f1c80d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "a26971a6856f41b2f796eb6bb501be0ef91ae9a435bea013e8658dd6cc4fa587", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "098aa420113a93bd88a022ce049fd85079942cba9bbab80913d5b8dd3495ec2d", "generation_seconds_batch": 18.562955867033452, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.197352409362793, "genuineness": 4.262581825256348, "identity": 0.6102809906005859, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer Vermoel die nieuwe Wandy doet.", "wer": 0.2631578947368421, "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_AUTH_reg/a26971a6856f41b2f796eb6bb501be0ef91ae9a435bea013e8658dd6cc4fa587.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "a2e3fcc4c60b4771078765348c94c498ab2aa3570c4f8257ce0c1ab723759840", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2fcf4ebaa3edc001bf39e45531f9727fc013ae1ab6918ae6e7c56bb67e342f5", "generation_seconds_batch": 26.495877828099765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 5.895530700683594, "genuineness": 4.042920112609863, "identity": 0.6106383204460144, "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_AUTH_reg/a2e3fcc4c60b4771078765348c94c498ab2aa3570c4f8257ce0c1ab723759840.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "a30abcf0beb90c8606d638d516755dd8e2f3ed70bfe89e9d91fdad1950cb5484", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a8e655f364edffac2201c807daa105ab761c2db8703791f0da60a1a77ae7d98", "generation_seconds_batch": 27.92258878494613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.412689685821533, "target_z": 1.15624988166591, "ce": 6.125868320465088, "genuineness": 4.582659721374512, "identity": 0.7553064823150635, "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_AUTH_reg/a30abcf0beb90c8606d638d516755dd8e2f3ed70bfe89e9d91fdad1950cb5484.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a32437e9bb80cb8f4ff04fcb3a1b7e21a4486b72912e3bff6b0ce73c96093a0c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f03b2a61c83e81d78a36cf579f52ed533b7a0aaffc8221ef215c561f4e17b257", "generation_seconds_batch": 25.82873369101435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.761687755584717, "target_z": 1.4921876550893756, "ce": 6.284721851348877, "genuineness": 4.68307638168335, "identity": 0.6143274903297424, "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_AUTH_reg/a32437e9bb80cb8f4ff04fcb3a1b7e21a4486b72912e3bff6b0ce73c96093a0c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "a3d7e064af7bacc207f62056e5f17d9b2bf10212638d56c5dd87f6138380bf89", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05179ef7b07fc820a4c5cad47e3bb6f184f35f70d8a15e5b335832d2f4b129e7", "generation_seconds_batch": 25.080744503997266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.035935401916504, "target_z": 2.718750075303512, "ce": 6.2767791748046875, "genuineness": 3.998987913131714, "identity": 0.5589914321899414, "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_AUTH_reg/a3d7e064af7bacc207f62056e5f17d9b2bf10212638d56c5dd87f6138380bf89.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a3e332e28919d6985e14bd120ebab4b50bb408745e309fa6dcbcd35793ae0215", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6149deabf8de5902ad1a82689493dd034797705b5ce4047a3c4fe72a188e7d9b", "generation_seconds_batch": 25.081363205914386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.086336612701416, "target_z": 1.8046874184211954, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.6823945045471191, "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_AUTH_reg/a3e332e28919d6985e14bd120ebab4b50bb408745e309fa6dcbcd35793ae0215.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "a3f80c06b070a954c880e496c0d535d2ace71f569dd65bacc78362c402a48320", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f376d0f64a1ab4a5a0d8789fbac089451076a0eae2d34c3f964acbe70bccdbd", "generation_seconds_batch": 26.133559937006794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 5.863759994506836, "genuineness": 3.8044304847717285, "identity": 0.5945080518722534, "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_AUTH_reg/a3f80c06b070a954c880e496c0d535d2ace71f569dd65bacc78362c402a48320.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a4017e87fb8a146e210f0abe0fda057ddc60a1e7246f8ccc13d5bddcf2c21ccb", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "72d014b8791e1c16c53d47fac99768664b411146cec6ee63432e0f401d1fd81a", "generation_seconds_batch": 19.916043839999475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.23706579208374, "genuineness": 4.444586753845215, "identity": 0.6818735599517822, "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_AUTH_reg/a4017e87fb8a146e210f0abe0fda057ddc60a1e7246f8ccc13d5bddcf2c21ccb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a4d877240c7bb209f3799d82e993e7fdf93374f14bbdf47283df4249516a15e1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed77a734085220ededb1a6dca2dcf84918583e9ceff42f6321bc2a34665999ab", "generation_seconds_batch": 25.30643321503885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.102569103240967, "target_z": 1.82031245248707, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.6721619367599487, "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_AUTH_reg/a4d877240c7bb209f3799d82e993e7fdf93374f14bbdf47283df4249516a15e1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "a54ee8e3f978bf5c499fe28e7feb00da129e87f3cfe50be0bc84806e0e45d5ff", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "028cd5b923470dff91e29b8b228ecb26c017c88fd96cfe5793a94743cdd87dae", "generation_seconds_batch": 19.347340421983972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.181467056274414, "genuineness": 4.193545341491699, "identity": 0.7889981269836426, "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_AUTH_reg/a54ee8e3f978bf5c499fe28e7feb00da129e87f3cfe50be0bc84806e0e45d5ff.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "a5e6850dbfbb9ef3be9f139bdb06b2c328562ff0b3988c5f76e8a17e040bcf9f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43f547c938c0ccc4e5c486b3a9acad9b67f463ccf5811ea531c4cb8fcb9e7aac", "generation_seconds_batch": 24.26740481692832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.889193058013916, "target_z": 0.6523436805235456, "ce": 6.38003396987915, "genuineness": 4.507347106933594, "identity": 0.4191622734069824, "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_AUTH_reg/a5e6850dbfbb9ef3be9f139bdb06b2c328562ff0b3988c5f76e8a17e040bcf9f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "a62bbf02eb2492f1ba512d91b98eec2031d7e392e80b2ece147dd9cc4687eccf", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0db5071a7b851e6ae8d958baa5dbfcf4cc77c6fbbb52cdb903b53754b8b770a7", "generation_seconds_batch": 18.461136849946342, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06198787689209, "target_z": 1.7812498673223838, "ce": 6.41180419921875, "genuineness": 4.331617832183838, "identity": 0.23314660787582397, "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_AUTH_reg/a62bbf02eb2492f1ba512d91b98eec2031d7e392e80b2ece147dd9cc4687eccf.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "a65c582ac0899e4f87df58c06f262c174df839dfd9472b54f81e4a8e40056561", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "47c6d267e1bdd6d84d5dc1b5a72a5e73cabc66d2b6a81b540d9301be78f5732f", "generation_seconds_batch": 18.6703474749811, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.022613525390625, "genuineness": 4.268857479095459, "identity": 0.6844954490661621, "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_AUTH_reg/a65c582ac0899e4f87df58c06f262c174df839dfd9472b54f81e4a8e40056561.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a69e6143f45cf9af6608421319c4b1f677e8281185c18d3bab432802a8e0b9b9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "318855408cc779b28aab940d623cf0afaa96a1c9e0449597296a28e535a5c45c", "generation_seconds_batch": 17.86445732298307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1837315559387207, "target_z": 1.898437622816442, "ce": 6.125868320465088, "genuineness": 4.250029563903809, "identity": 0.6756318807601929, "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_AUTH_reg/a69e6143f45cf9af6608421319c4b1f677e8281185c18d3bab432802a8e0b9b9.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": "a6e439d3a30a36cb833a31f14afe25fcabe02fde2c8ddfdf77f2cc76a719857a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9519bb1b7b0d426caa083eef5e106140f039a0765fcbc759c8e6310bce52f1b", "generation_seconds_batch": 16.46590609604027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.308549880981445, "genuineness": 4.262581825256348, "identity": 0.5992951989173889, "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_AUTH_reg/a6e439d3a30a36cb833a31f14afe25fcabe02fde2c8ddfdf77f2cc76a719857a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "a6fc7924addd01e6fc1a90dcc03c7da3ee739469e82928aed7cb44be463f35b0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc93e2cb68f1b140d12f806d9fb93da4f6b8ff3bb2e5d13f975c33b996c30871", "generation_seconds_batch": 20.897666932898574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.078212261199951, "genuineness": 4.406930446624756, "identity": 0.4695788025856018, "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_AUTH_reg/a6fc7924addd01e6fc1a90dcc03c7da3ee739469e82928aed7cb44be463f35b0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "a712b3431a69576a8f1128577bdedf23913c26ffde7cd2645a776dae7f39a7f6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d3e6c7f1394dcee72706e1af86a3e3c3d3b663535f8c2aff9c449c565beb495", "generation_seconds_batch": 27.49681421497371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1982598304748535, "target_z": 2.874999956969422, "ce": 5.855817794799805, "genuineness": 3.9173994064331055, "identity": 0.6922561526298523, "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_AUTH_reg/a712b3431a69576a8f1128577bdedf23913c26ffde7cd2645a776dae7f39a7f6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "a79f66237623a2b4b78088c1c9f331484e0548554ef07d11c01cd008a71c35e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32fbd604149c126bdd1a358644a100112b8b4f0c0ba7cf26a5f42e97b3df826e", "generation_seconds_batch": 25.280982254073024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.7329863905906677, "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_AUTH_reg/a79f66237623a2b4b78088c1c9f331484e0548554ef07d11c01cd008a71c35e5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "a826bd6b19f29099e2e6911410deebd58be7c5dc4c9ff35fa97b961d111393a6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3a3a25522f0a1a2d43e2d59f068fc8e16915f0d6635b0b21e74129d4b3cec59", "generation_seconds_batch": 19.932358084013686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 6.062326908111572, "genuineness": 4.319066047668457, "identity": 0.6033706665039062, "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_AUTH_reg/a826bd6b19f29099e2e6911410deebd58be7c5dc4c9ff35fa97b961d111393a6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "a835dc88204cd1c5cb32ae6c45738bdff97520b51e5eae022a6f00be6614db18", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caa52d9c699cc8f8a445e6d177d088e7ce04bd1ef86d698cef94d0b96d389d92", "generation_seconds_batch": 29.06020836799871, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.4912309646606445, "genuineness": 4.281409740447998, "identity": 0.5813445448875427, "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_AUTH_reg/a835dc88204cd1c5cb32ae6c45738bdff97520b51e5eae022a6f00be6614db18.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "a84e911a0f3311806795be789145e10e6df186395da63bf9b9c8c146a5b17f1a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95387ad9d15c0a8fc3b5b94f61215baf5cf29e4fe3eee804fbed28ad5fafcf1a", "generation_seconds_batch": 25.83667141397018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.260893821716309, "genuineness": 4.18099308013916, "identity": 0.746414065361023, "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_AUTH_reg/a84e911a0f3311806795be789145e10e6df186395da63bf9b9c8c146a5b17f1a.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "76d8ad10ca5d7ff1924ec05ed462f39784034632d90ade9a00676d4eb73a0bb0", "generation_seconds_batch": 20.0404176939046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.173524379730225, "genuineness": 4.281409740447998, "identity": 0.5208273530006409, "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_AUTH_reg/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "0bea0b5417db57d69e53cec656b49653a8c1a179dc9f61d7d7a70b7b1e74affa", "generation_seconds_batch": 18.714035636978224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4208059310913086, "target_z": 1.1640623986988472, "ce": 6.09409761428833, "genuineness": 4.3127899169921875, "identity": 0.6555029153823853, "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_AUTH_reg/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "a9e45e15afcc7887a158baccd1dcc0849d6bfed429080eb55eea8f44d3f42d45", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5be2f1b278425fb3f4d8ed6bd9070ed4e80bb0fe596486d6736868328e3c6768", "generation_seconds_batch": 26.548717948026024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.0305562019348145, "genuineness": 4.030367851257324, "identity": 0.626177191734314, "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_AUTH_reg/a9e45e15afcc7887a158baccd1dcc0849d6bfed429080eb55eea8f44d3f42d45.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "aa01aba241deefddabf0ebb22ca3dfea4ebf68ebe5b2b84a7ac48c60f2800c0c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a20d54024598d34d7415f0645fc294052381f577c89931e4c813198a2c762f2e", "generation_seconds_batch": 25.960712919011712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0944528579711914, "target_z": 1.8124999354541327, "ce": 6.181467056274414, "genuineness": 4.268857479095459, "identity": 0.6943938136100769, "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_AUTH_reg/aa01aba241deefddabf0ebb22ca3dfea4ebf68ebe5b2b84a7ac48c60f2800c0c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "aa96195cd43b645bf40dee99cbde41bb3336a562d7e5d9e260ee38be6ec6e761", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "555c41b694875bb49dabad8d4037c996502a8ddac3c9764d3b157587ca791a07", "generation_seconds_batch": 25.442441481980495, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.332377910614014, "genuineness": 3.9801597595214844, "identity": 0.3352845013141632, "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_AUTH_reg/aa96195cd43b645bf40dee99cbde41bb3336a562d7e5d9e260ee38be6ec6e761.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "aa972283278d0be6f32d4447adb218dbb6c47ca2d2655bbee4a39060ebfa741b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf4044602fcfc07930dd9d75a055292e811370f88029683118bd68ddd3a569f5", "generation_seconds_batch": 24.453199386945926, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.133810997009277, "genuineness": 4.093128204345703, "identity": 0.37491142749786377, "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_AUTH_reg/aa972283278d0be6f32d4447adb218dbb6c47ca2d2655bbee4a39060ebfa741b.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "52b238e69ed02f8cb003bc33c79e12ac4c47fe4398ba5c643de248586e249f81", "generation_seconds_batch": 16.35831957194023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.403861999511719, "genuineness": 4.118232727050781, "identity": 0.28507375717163086, "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_AUTH_reg/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "abf9b3e4519df1f891569daf031569da9ff934b74185ca35efea9fbe01696b8e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9fd2d66486a49422373725dfc8b5c6c955fe361cbab8cba64738aec110e9f0a", "generation_seconds_batch": 24.002248284989037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 6.062326908111572, "genuineness": 4.13078498840332, "identity": 0.6600240468978882, "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_AUTH_reg/abf9b3e4519df1f891569daf031569da9ff934b74185ca35efea9fbe01696b8e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "ac99c8ac7e1c15b7c1609a9925d677cfc6b7e2051fb26d945cd568470fb3ca21", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34fe1b2effb012cae36e2e1d50dd5868c1c36cb8948dd95428c949d65a47ca6f", "generation_seconds_batch": 19.131115519907326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.300607204437256, "genuineness": 3.9801597595214844, "identity": 0.5970275402069092, "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_AUTH_reg/ac99c8ac7e1c15b7c1609a9925d677cfc6b7e2051fb26d945cd568470fb3ca21.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "ad3e7fc70a86d0d42c22063a6459591fab2800d81b16f359bfe3ba6f28ddf279", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "841411f15ea1e54d01be466c036cbcc89e221fc75b64b692bdb0fbe2c0f37aa5", "generation_seconds_batch": 28.671361761982553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0684003829956055, "target_z": 2.7500001434352606, "ce": 6.165581703186035, "genuineness": 4.18099308013916, "identity": 0.553983211517334, "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_AUTH_reg/ad3e7fc70a86d0d42c22063a6459591fab2800d81b16f359bfe3ba6f28ddf279.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "ad7d096eeeec926ffb7d597e53ef9e78b99c3957a3bf6129109555a05ca96981", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1cf21d9ac1de0cfe189a6c4993827222f5694bdcf46fd4d46ec541407e32f39", "generation_seconds_batch": 19.004660979961045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0376393795013428, "target_z": 1.7578125457199893, "ce": 6.133810997009277, "genuineness": 3.9550557136535645, "identity": 0.6863260269165039, "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_AUTH_reg/ad7d096eeeec926ffb7d597e53ef9e78b99c3957a3bf6129109555a05ca96981.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "6f932d90ece97290b370ceeda7daf79f5bf2b8699cfba02c4fa01f8ce65d332f", "generation_seconds_batch": 18.61614672304131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3315277099609375, "target_z": 1.0781251703293722, "ce": 6.1894097328186035, "genuineness": 4.469690799713135, "identity": 0.6019692420959473, "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_AUTH_reg/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.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": "adfd014e669954530f8b2db4348c3afa7947e2cb2f99326b6b31f41f7136990d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bb2c1abe739eb9956ed59adc0813374ad41a6d859353089f293430f0f0c4f3c", "generation_seconds_batch": 16.732476220000535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2892422676086426, "target_z": 1.9999998852517915, "ce": 6.205295085906982, "genuineness": 4.3127899169921875, "identity": 0.7331342101097107, "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_AUTH_reg/adfd014e669954530f8b2db4348c3afa7947e2cb2f99326b6b31f41f7136990d.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "a77a13db23bfd85818924a2a67071fbd5a111fca8525726a0567df233444bea7", "generation_seconds_batch": 20.615297571988776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4353346824645996, "target_z": 2.140625191844661, "ce": 6.38003396987915, "genuineness": 3.9801597595214844, "identity": 0.4021204710006714, "hypothesis": "The road turns left behind the bridge. Keep going until you reach this small wooden gate. I will meet you on the other side.", "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_AUTH_reg/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.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": "ae6efc4edd0c15dc0ed6f2f36ab03bfa500461c0b64ee7294dd5bba8b9eb420d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81a54a458ad3877c1af99daff67aba04c232d22dc6d70c5f996065f5331b1fa6", "generation_seconds_batch": 16.544586537987925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.054384231567383, "genuineness": 4.074300289154053, "identity": 0.6950271725654602, "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_AUTH_reg/ae6efc4edd0c15dc0ed6f2f36ab03bfa500461c0b64ee7294dd5bba8b9eb420d.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": "aeeea6adb6e5549f2abe584b4fca91f5706fac37264cd4e82e4d607b7278419c", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a96ccec72cdd05a126e41ce134099f7742aa3c89c129dd9f7d5ebf4e731fa5bd", "generation_seconds_batch": 26.45291008404456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9872379302978516, "target_z": 2.6718749731058886, "ce": 6.24500846862793, "genuineness": 4.256305694580078, "identity": 0.582840085029602, "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_AUTH_reg/aeeea6adb6e5549f2abe584b4fca91f5706fac37264cd4e82e4d607b7278419c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "af072e504b138c6d8ba544a69338c112beea46c60bf43aa9586289c55284cc8b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "445e52b3c85ce4ee9e4e5a54db22eb1086f0407ef7dceb5387c5a402ed113406", "generation_seconds_batch": 25.902048235060647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.141753673553467, "genuineness": 4.011539936065674, "identity": 0.6779474020004272, "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_AUTH_reg/af072e504b138c6d8ba544a69338c112beea46c60bf43aa9586289c55284cc8b.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": "af0cbfae75c3a77297e1780640dddef672a6b1910c42110fd62ac3e20d5f2c31", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca11eed5c563f6ebbee65d7d67b6cb0aaebb6e21affa1b5e9dea512d3f7ecfa4", "generation_seconds_batch": 22.143539529992267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.252951145172119, "genuineness": 4.20609712600708, "identity": 0.4124521017074585, "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_AUTH_reg/af0cbfae75c3a77297e1780640dddef672a6b1910c42110fd62ac3e20d5f2c31.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "af424f5a1c00f5952d779236204824b7c798c7080d9bc6fb0fbabef6da335192", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "862f0ab462792d6eccff541e7f5b0010655c20e9911f1c9f917c2c66ec04b51a", "generation_seconds_batch": 24.22340861300472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.38003396987915, "genuineness": 4.055472373962402, "identity": 0.4397512674331665, "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_AUTH_reg/af424f5a1c00f5952d779236204824b7c798c7080d9bc6fb0fbabef6da335192.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "af689acb6ff8bf8d54ef03e29b806bded46c2c57c7cfc2d34e226b8d5be5fd4c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1561239444bca8114e579a86c92dead06acc21001c9247ff0b8a8fea0dc65fa5", "generation_seconds_batch": 24.418968347017653, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.157639026641846, "genuineness": 4.124508857727051, "identity": 0.2873769998550415, "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_AUTH_reg/af689acb6ff8bf8d54ef03e29b806bded46c2c57c7cfc2d34e226b8d5be5fd4c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "afc0b8a46768a2beb09519364c7dad7b8b814d875cb1dd70aaf60ccdefbe8b63", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "074c52a7b5a979292123628832864f31601ffb8c26002c29d5d0fd9f3a6d3a1b", "generation_seconds_batch": 27.705699809943326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.268836498260498, "genuineness": 4.356721878051758, "identity": 0.7221372127532959, "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_AUTH_reg/afc0b8a46768a2beb09519364c7dad7b8b814d875cb1dd70aaf60ccdefbe8b63.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "afc535c43ff4fb54b40d29fe3b9a83ea7dd61177e360feccdf70fa3b28d2766f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b03bae7d3145b011c814282b60694f1b4dbe4d30aa8c7884df239a53a0f1e461", "generation_seconds_batch": 17.296098595950752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.630124092102051, "target_z": 2.32812514164232, "ce": 6.340320587158203, "genuineness": 4.381826400756836, "identity": 0.6353167295455933, "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_AUTH_reg/afc535c43ff4fb54b40d29fe3b9a83ea7dd61177e360feccdf70fa3b28d2766f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "b022ba8d90f6ca9501badb44ac2dc72ee5279e0d2372e1f98c7ac725a262eb6b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "963a1e625465ecab96a61d7d94dd7923ec4961077e4fa220b4acf5847585056e", "generation_seconds_batch": 18.400722349062562, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9077796936035156, "target_z": 1.6328125026894111, "ce": 6.300607204437256, "genuineness": 4.331617832183838, "identity": 0.7743439674377441, "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_AUTH_reg/b022ba8d90f6ca9501badb44ac2dc72ee5279e0d2372e1f98c7ac725a262eb6b.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": "b0726dae460f35d97a8a807f30d55ac0873f86e2c3ffbf59bf35c6b76930e20b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e64f99a4f016a0275980a2c3ebda6daf2333718d75f063a352caec22a2cd9e99", "generation_seconds_batch": 24.24600269505754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.181467056274414, "genuineness": 4.118232727050781, "identity": 0.48361828923225403, "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_AUTH_reg/b0726dae460f35d97a8a807f30d55ac0873f86e2c3ffbf59bf35c6b76930e20b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "b08395a053b718a96fdd30e4ddeea95312d29919794fab7eb85768e1bb710986", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c3de2a7a5bb2bf98e20d65e4fa4147b19dddc276deb416dcf78e223d220a94f", "generation_seconds_batch": 26.691403631935827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 6.332377910614014, "genuineness": 4.369274139404297, "identity": 0.16282647848129272, "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_AUTH_reg/b08395a053b718a96fdd30e4ddeea95312d29919794fab7eb85768e1bb710986.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "b0eedfce5369a675a2d034f1a78eaacdc3f9c1b391e177540589ec6346599224", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b3e9ccba619c399aad890362198cdcbda1fafae7581678e622cf438a6ed36e9", "generation_seconds_batch": 24.53017605491914, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.6308244466781616, "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_AUTH_reg/b0eedfce5369a675a2d034f1a78eaacdc3f9c1b391e177540589ec6346599224.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": "b1827c5b3c68916f287e40f3793abe53affae0cba0c6c691025ab4fbf96a72db", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8daeb76262b733f6390c466f420a0c85be8f86412ab9dac0d721954aa96d22ff", "generation_seconds_batch": 16.58228388905991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13332986831665, "target_z": 2.812499820705924, "ce": 6.125868320465088, "genuineness": 4.005263805389404, "identity": 0.49231430888175964, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/vn_S_AUTH_reg/b1827c5b3c68916f287e40f3793abe53affae0cba0c6c691025ab4fbf96a72db.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": "b1a8aef8f32301f15c9ef1a150493f5a6d635844fcda5cb171e0db0d33237dc5", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0c1ce1d92bedde9d7b4e5d77c12f61e8c50d79347d132bcee1ce86b75c21b74", "generation_seconds_batch": 25.918430355959572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.284721851348877, "genuineness": 4.005263805389404, "identity": 0.6502914428710938, "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_AUTH_reg/b1a8aef8f32301f15c9ef1a150493f5a6d635844fcda5cb171e0db0d33237dc5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "b1e06faf2b59a7326e0b62d965fedd71f8699bbbd69761e2cf0071a5ff4bc218", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8332a1d60683b61a823ce865fdec20579249bed74e53a8ca1876ace78fadefd", "generation_seconds_batch": 25.451868592062965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9223079681396484, "target_z": 2.609374836842391, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.6289387345314026, "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_AUTH_reg/b1e06faf2b59a7326e0b62d965fedd71f8699bbbd69761e2cf0071a5ff4bc218.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "b25f8550c3f9b88eb30922c517c1ebf1037cc1d08a30198b357cc70e4871ebd2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "225bbb0463bbc012cb844ef7bcbb66f5e9751c4936c5eb402182300dc7ea94d8", "generation_seconds_batch": 25.83667141397018, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.117925643920898, "genuineness": 4.15588903427124, "identity": 0.663348913192749, "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_AUTH_reg/b25f8550c3f9b88eb30922c517c1ebf1037cc1d08a30198b357cc70e4871ebd2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "b268ad256ec9a6fe5bc675d85c85309bc603c4593af3d76ead6d99209e2e9508", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f765ee0e0132d86fdb96c27a252ce185581e9f164df4bab42d3a2dadbb5c86ff", "generation_seconds_batch": 19.287579694064334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.956477165222168, "target_z": 1.6796876048870344, "ce": 6.419747352600098, "genuineness": 4.394378662109375, "identity": 0.5724489092826843, "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_AUTH_reg/b268ad256ec9a6fe5bc675d85c85309bc603c4593af3d76ead6d99209e2e9508.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "b2bc4bd18359929fa6f8cf2c4ab178815570be519a2c900f0c2d2eee99ec86b6", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "ebfd3f28dbb5eed364df314f1b658a6d193ba4d20dfffc960549ec955c44369f", "generation_seconds_batch": 20.349048105999827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.149696350097656, "genuineness": 4.319066047668457, "identity": 0.632438063621521, "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_AUTH_reg/b2bc4bd18359929fa6f8cf2c4ab178815570be519a2c900f0c2d2eee99ec86b6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "b3053ec18d9f0372ec88ffcad0a72f7b817c8433bdd0f4e117671a550a3a4e4c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ec47be3e7dab71d2226b1d259efec9fd320a936eed3b2f49ec8a0e03d7ce69e", "generation_seconds_batch": 25.923318175948225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.2767791748046875, "genuineness": 4.042920112609863, "identity": 0.39361992478370667, "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_AUTH_reg/b3053ec18d9f0372ec88ffcad0a72f7b817c8433bdd0f4e117671a550a3a4e4c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "b32f7cf0efb593a397c3f6b018ab63da86226eb0b097b742152fd4e816a03638", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d456c68ea6942321de9f90a248711664897639251fe36234bba8bd0d4931e84b", "generation_seconds_batch": 27.508233027998358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.95477294921875, "target_z": 2.64062490497414, "ce": 5.7128496170043945, "genuineness": 4.042920112609863, "identity": 0.8247665166854858, "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_AUTH_reg/b32f7cf0efb593a397c3f6b018ab63da86226eb0b097b742152fd4e816a03638.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "b3809e8811361f3602f32a02e5e5211c229aa0fe5963e9395b17c7e3d24bdec1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89f44e7c21f7a0d1c77a068a52da0757dba2ec6165a3f1ff63beabe323ef5f1a", "generation_seconds_batch": 25.52621907205321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.193551540374756, "target_z": 0.9453123807694396, "ce": 6.340320587158203, "genuineness": 4.2751336097717285, "identity": 0.4804002046585083, "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_AUTH_reg/b3809e8811361f3602f32a02e5e5211c229aa0fe5963e9395b17c7e3d24bdec1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "b3dd0da68e8554c5d14a46392954bdd2cfe3e6e096468c74b9f58443c38fc034", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "007d06ce357f023549c9b1752e58af901e3e833be5931456f7eca0437004c310", "generation_seconds_batch": 24.752158147050068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6561765670776367, "target_z": 1.390624933661192, "ce": 6.09409761428833, "genuineness": 4.042920112609863, "identity": 0.7241668701171875, "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_AUTH_reg/b3dd0da68e8554c5d14a46392954bdd2cfe3e6e096468c74b9f58443c38fc034.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "cc6cf39f94ef45fcab2381793d978ad53dc7d9f25e780ad58bd232adaa50b7ba", "generation_seconds_batch": 17.86702163692098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3964571952819824, "target_z": 1.1406248476000356, "ce": 6.268836498260498, "genuineness": 4.319066047668457, "identity": 0.5070626139640808, "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_AUTH_reg/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "b47bb7737e5fb79959dadb82db61a60d27f1f93b4bc96347d127f7090ee92cae", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e06a63b4ca42529e01df49a7609bfda55eabcbf6c20bae3c0fbbca0e07f55755", "generation_seconds_batch": 24.48547170497477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.6531229615211487, "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_AUTH_reg/b47bb7737e5fb79959dadb82db61a60d27f1f93b4bc96347d127f7090ee92cae.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "b49cfa5691cf78ca420179d54c07e57cd777ed15e59f1bb18e792263570098e5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37563a134afd1dcdfbea6c7f843854a4a725f3177f97e465dc44e4e51ec965f9", "generation_seconds_batch": 24.894474033964798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.0305562019348145, "genuineness": 4.068024158477783, "identity": 0.5463504791259766, "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_AUTH_reg/b49cfa5691cf78ca420179d54c07e57cd777ed15e59f1bb18e792263570098e5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "b4b5464f35ec5e1fb15036805f359610e4ce65dd8c80b7924ef652b3858ba214", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d8197c401a99281e4cd103079761804f2fed1eba5814300582e37d61975213ac", "generation_seconds_batch": 25.7473061659839, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.181467056274414, "genuineness": 4.243753433227539, "identity": 0.6464604735374451, "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_AUTH_reg/b4b5464f35ec5e1fb15036805f359610e4ce65dd8c80b7924ef652b3858ba214.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "b4c70a8283e3f62bce86ea18614faad471510cf9d476a08b44ef91dc03a3779f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98d66b650ea3682619462b4702d20c4aacde1eb9c477f493ad22a143212f6b28", "generation_seconds_batch": 26.648954247008078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.850965976715088, "target_z": 1.5781248834588506, "ce": 6.109982967376709, "genuineness": 4.111956596374512, "identity": 0.7230846881866455, "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_AUTH_reg/b4c70a8283e3f62bce86ea18614faad471510cf9d476a08b44ef91dc03a3779f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "b4d03c1120a3f3bce13b4b01b6f56a9dd39e8189a5669512ee7b5c696c5b5afd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0e08fa9df0d2f31cd985fcf094d1069d076d93646ed402470ff3cf07cfbac99", "generation_seconds_batch": 25.959222548059188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.284721851348877, "genuineness": 4.381826400756836, "identity": 0.5347232818603516, "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_AUTH_reg/b4d03c1120a3f3bce13b4b01b6f56a9dd39e8189a5669512ee7b5c696c5b5afd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "b4f05ad323b1b4dffd29b0079f75717a4ddedcf9255280e5c0db17e7b29863e9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ec9609969e34f7c2bc797f90204d43cac9a135a03a4b6c9fb607e5ba28d7e60", "generation_seconds_batch": 19.467890167958103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.062326908111572, "genuineness": 4.394378662109375, "identity": 0.546333372592926, "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_AUTH_reg/b4f05ad323b1b4dffd29b0079f75717a4ddedcf9255280e5c0db17e7b29863e9.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": "b51c7b5321953307785b189c61afdc05e450f82974aff8ff8daf057aa2add76c", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab34889d782770733069d66b1c32af1f1031622998c94cc7fd977ebbe010ff9b", "generation_seconds_batch": 22.408172048046254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.151266574859619, "target_z": 1.8671875546846932, "ce": 6.38003396987915, "genuineness": 4.18099308013916, "identity": 0.3480377793312073, "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_AUTH_reg/b51c7b5321953307785b189c61afdc05e450f82974aff8ff8daf057aa2add76c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "b51ebb2a82f72b3cd5867c7b807be8100fb2d1c3eaadf4be9973eec349ad9a25", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "815cab5c0eecd7782314cd7f0113e22d375296f110b7f8472f108ef543ea999d", "generation_seconds_batch": 29.17227515799459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.268836498260498, "genuineness": 4.281409740447998, "identity": 0.5577284693717957, "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_AUTH_reg/b51ebb2a82f72b3cd5867c7b807be8100fb2d1c3eaadf4be9973eec349ad9a25.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "b5437009524271252da1db9db363d7ec6deb6e1b7eb21c8aff8da3274ec0a21a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6febf10dfe170a8adcc65e09e50410b5c16f9de55e153c258aca7d263175b77", "generation_seconds_batch": 27.92258878494613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.165581703186035, "genuineness": 4.243753433227539, "identity": 0.688535213470459, "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_AUTH_reg/b5437009524271252da1db9db363d7ec6deb6e1b7eb21c8aff8da3274ec0a21a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "b58d467f84286d8c2246e7fdfb9ddceeb8a26a00b8f9ef59092aa2833bd0e95e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44f1bac0be5a5757785873e058ba44513999441a2b56c6f20e35f5306e36ba90", "generation_seconds_batch": 24.2462028299924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.284721851348877, "genuineness": 4.532451152801514, "identity": 0.6621742844581604, "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_AUTH_reg/b58d467f84286d8c2246e7fdfb9ddceeb8a26a00b8f9ef59092aa2833bd0e95e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "b6a67f28cd2c2bbec8d20c13b968954d413f9d93df5ecf21c98dbea366b9d0d9", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8cff49c40eeb81c4d3f81de29907e39690ea9faf7b43ae5177a90db827b339d", "generation_seconds_batch": 25.209855598048307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.124563694000244, "target_z": 0.8789062154858904, "ce": 6.165581703186035, "genuineness": 4.406930446624756, "identity": 0.6802442073822021, "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_AUTH_reg/b6a67f28cd2c2bbec8d20c13b968954d413f9d93df5ecf21c98dbea366b9d0d9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "b6c65a3783747cdf938830e68641ad111a06b16f37510188ba666e424dd23c2e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "283d28d108105e9c6341bfd54f9de2e86eb798485b967039971786804411a814", "generation_seconds_batch": 20.626020450959913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.213237762451172, "genuineness": 4.432034492492676, "identity": 0.7244604825973511, "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_AUTH_reg/b6c65a3783747cdf938830e68641ad111a06b16f37510188ba666e424dd23c2e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "b746f6bd76fa58c80c208034102fb9f06c017a3ae06830f003907c11624bd13a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d0be2b715d6c13e6c30d6f216bebeb1d35d797055bc021e701f3d477e5d5bc79", "generation_seconds_batch": 19.99853840400465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5425496101379395, "target_z": 1.2812501541929053, "ce": 6.173524379730225, "genuineness": 4.381826400756836, "identity": 0.6242091655731201, "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_AUTH_reg/b746f6bd76fa58c80c208034102fb9f06c017a3ae06830f003907c11624bd13a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "b8330979109b86b1ae8853fb5893468793421211c0a544a3643fe611514aeb3c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be40329332ba4e116ee23a7bc2778ae01f7e282143748cf950a86b1096420354", "generation_seconds_batch": 26.7274280450074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.295654773712158, "target_z": 2.9687501613646683, "ce": 6.340320587158203, "genuineness": 4.143336772918701, "identity": 0.5584429502487183, "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_AUTH_reg/b8330979109b86b1ae8853fb5893468793421211c0a544a3643fe611514aeb3c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "b874f19bc8600a0d3f9914b0a896cfb62019fdfba33b464dfc3f719033878c30", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0131657391de815355c7838e3813a08ebd1c7e06c77c2fa88816b5e0616a8a5a", "generation_seconds_batch": 26.70499136694707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.300607204437256, "genuineness": 3.8671910762786865, "identity": 0.6013160943984985, "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_AUTH_reg/b874f19bc8600a0d3f9914b0a896cfb62019fdfba33b464dfc3f719033878c30.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": "b8976a425cee9a659ea20ee2df2df754afe33d03bd8235364b2b767e2c08f44a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e96b55f2a824290ef996578a280b962889038136c97fcf3f458febbbfdeb89de", "generation_seconds_batch": 19.739179227966815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.419747352600098, "genuineness": 4.030367851257324, "identity": 0.2636057734489441, "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_AUTH_reg/b8976a425cee9a659ea20ee2df2df754afe33d03bd8235364b2b767e2c08f44a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "b8aa6330ebfac706f5cf6ffa4cac29238a3fadf4afeda864a6334f54ffd2f06d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b43aa2cd9488daf94f110b5a4e57c3e3188ce44309e1cb41e6b2d7971dd2bd4d", "generation_seconds_batch": 25.83938662696164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.181467056274414, "genuineness": 4.306513786315918, "identity": 0.6475392580032349, "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_AUTH_reg/b8aa6330ebfac706f5cf6ffa4cac29238a3fadf4afeda864a6334f54ffd2f06d.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": "b8cc6850b1ebd8450326363625c99ee7561df4822ffdf7287156001ac1a95e43", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe6df6ebfcb7e870a91b9c348a2b5e8a01dece3913414c1190df5540b9904921", "generation_seconds_batch": 16.752967583946884, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.316492557525635, "genuineness": 3.8044304847717285, "identity": 0.5240938663482666, "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_AUTH_reg/b8cc6850b1ebd8450326363625c99ee7561df4822ffdf7287156001ac1a95e43.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "b9a2747baeab1c9d5879e6fff3245d05cd05afa52e441c6d42e9016e0b82ff99", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2baa4b9e5600cabcbdb84bb54ab1e513b359af00a993d8962a274ac265130a60", "generation_seconds_batch": 26.041506521985866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.759983539581299, "target_z": 2.453124955176481, "ce": 6.443574905395508, "genuineness": 4.243753433227539, "identity": 0.6213697195053101, "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_AUTH_reg/b9a2747baeab1c9d5879e6fff3245d05cd05afa52e441c6d42e9016e0b82ff99.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "b9f98387982800edf271a42fafc8a6052c10f419a6381dcb6fe928ab383bd5b8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2ad364719d4a0ba51366d7c45771d595af3082a44e4b61185f3cee4863a95169", "generation_seconds_batch": 24.270862029050477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3964571952819824, "target_z": 1.1406248476000356, "ce": 6.1020402908325195, "genuineness": 4.444586753845215, "identity": 0.7739251255989075, "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_AUTH_reg/b9f98387982800edf271a42fafc8a6052c10f419a6381dcb6fe928ab383bd5b8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "ba17e7c35e42558911d48b99b9c41792d993d121cd3d0bd099300750d3a02e51", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5012038f33c8844f5ad8930bf7c5221b6586d6c0a1ea6424c53cc16d2d995735", "generation_seconds_batch": 25.902048235060647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.23706579208374, "genuineness": 4.306513786315918, "identity": 0.75386643409729, "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_AUTH_reg/ba17e7c35e42558911d48b99b9c41792d993d121cd3d0bd099300750d3a02e51.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "ba354429e9839be70b482ac9a4231515b3afc9cdeb796295266b3319bdee3d2d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f393955a29ac3f6fbd512156edaa6ef47c186de3680180243a7838a72703a42b", "generation_seconds_batch": 27.640296171070077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.213237762451172, "genuineness": 4.331617832183838, "identity": 0.48946908116340637, "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_AUTH_reg/ba354429e9839be70b482ac9a4231515b3afc9cdeb796295266b3319bdee3d2d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "ba43a8d8433ec144277d89a5d66e25ee4e25cab5c89a851f0a16bab504a5877b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3708d4ddc89cdc1b2e216758a5ea9d52a8e021f8909f95da4a4e4929c16a9c50", "generation_seconds_batch": 25.66594668000471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.332377910614014, "genuineness": 4.356721878051758, "identity": 0.4413014054298401, "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_AUTH_reg/ba43a8d8433ec144277d89a5d66e25ee4e25cab5c89a851f0a16bab504a5877b.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "e5e96bc6f58d8b661dacd2dd45f095d979188f16d482dee9e7f5a3c6f003644e", "generation_seconds_batch": 22.326247585006058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.339643716812134, "target_z": 1.0859374578658922, "ce": 6.165581703186035, "genuineness": 4.7081804275512695, "identity": 0.8089346289634705, "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_AUTH_reg/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "bb240c4516e95050cfb5ba5028f9665705f9edbfba722d28ef43ed11e5cbc99d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98f22a7387dd3b1a27be4b6a1171c8f4df553b6c0bc0e7025e6d87a3ecaeb0cb", "generation_seconds_batch": 25.046680395957083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.09409761428833, "genuineness": 4.124508857727051, "identity": 0.403895765542984, "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_AUTH_reg/bb240c4516e95050cfb5ba5028f9665705f9edbfba722d28ef43ed11e5cbc99d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "bb4b8d757ef9148bec8cb6e04701e8a8a0577d391f5222c8b14ad52b9c2d42c3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2df6255f3c060ff5a29204ea7b1d0087e83599a2cab9fedd801f887ddc050cec", "generation_seconds_batch": 19.844501759973355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.402869701385498, "target_z": 2.1093751237129124, "ce": 6.133810997009277, "genuineness": 4.124508857727051, "identity": 0.5897525548934937, "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_AUTH_reg/bb4b8d757ef9148bec8cb6e04701e8a8a0577d391f5222c8b14ad52b9c2d42c3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "bbaee42a9e70ea087f9fa9ed340da359c4fedd8290ca324d4c99a3f3c8039898", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6696c085d2fbd5e5e8b8b51654eec4fb0718b9767ab47e0efa14a04b2e561c60", "generation_seconds_batch": 23.91770193609409, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.332377910614014, "genuineness": 4.042920112609863, "identity": 0.7255113124847412, "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_AUTH_reg/bbaee42a9e70ea087f9fa9ed340da359c4fedd8290ca324d4c99a3f3c8039898.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": "bbc260ddeaa731441ec0db6135412fad38210ea248b25f857c1f67ee30eaf7ac", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c12feff9c1a2651629351503c1620a0d8e88124093a30fbea4f67b3a2b369995", "generation_seconds_batch": 16.61147071502637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.5723568201065063, "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_AUTH_reg/bbc260ddeaa731441ec0db6135412fad38210ea248b25f857c1f67ee30eaf7ac.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": "bc6559638b13d17eade66923201a717c037eac377107882f803a0bebe28742c9", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fae4e75ce5cc2e593e8b8cabb9ce8ffd5902b4b0dd75253f89c24bc73e7a0e9c", "generation_seconds_batch": 16.673505916027352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.197352409362793, "genuineness": 3.948779582977295, "identity": 0.7780969142913818, "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_AUTH_reg/bc6559638b13d17eade66923201a717c037eac377107882f803a0bebe28742c9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "bcbb4c1d45ce53e2b92822692923fa30a25b699fd87464a0adab4734104f0aad", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "781d116e4c426c56dea2752c38d7dcb9968d16e5ebef8158ed1024fbaedadd4b", "generation_seconds_batch": 25.90016617393121, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9872379302978516, "target_z": 2.6718749731058886, "ce": 6.395919322967529, "genuineness": 4.011539936065674, "identity": 0.45713356137275696, "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_AUTH_reg/bcbb4c1d45ce53e2b92822692923fa30a25b699fd87464a0adab4734104f0aad.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": "bcff149de491d6f3453ecf8914e7ad63b535fb26f506afd606f5b17cb53b7c30", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3d908111a440133007bcf945268b89914b650c7eb5615e34303f8b020b78e0e", "generation_seconds_batch": 19.763231093995273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.623711585998535, "target_z": 1.359374865529443, "ce": 6.427689552307129, "genuineness": 4.293961524963379, "identity": 0.3976227045059204, "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_AUTH_reg/bcff149de491d6f3453ecf8914e7ad63b535fb26f506afd606f5b17cb53b7c30.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "bd28d89e96a5c66c207cc7b44750db1acd5f3f39b7eb930c43901449c1c381c7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1038d58bcd73778ea989170d2da95c197dfd98288251d21588b8b11afed30857", "generation_seconds_batch": 24.752158147050068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5912468433380127, "target_z": 1.3281250268941114, "ce": 5.919358730316162, "genuineness": 4.369274139404297, "identity": 0.44986340403556824, "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_AUTH_reg/bd28d89e96a5c66c207cc7b44750db1acd5f3f39b7eb930c43901449c1c381c7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "be9dcb025e9ad57c647005ccbd4e3e2b3ac288930a99143e0fcf2350b9621879", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0265add83387360d9c5d6a647636e24c182e13ce9c440de36bae0a4569101e54", "generation_seconds_batch": 26.648954247008078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3802249431610107, "target_z": 1.1250000430305782, "ce": 6.125868320465088, "genuineness": 4.306513786315918, "identity": 0.780382513999939, "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_AUTH_reg/be9dcb025e9ad57c647005ccbd4e3e2b3ac288930a99143e0fcf2350b9621879.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": "befdb12e2a52ab98b364fbcfe1cdeab2205daed2dbfe801fbdb42da39d09ba1e", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f23570c707b374c05868a2976fb66432ae9099fa681e38b4687e089b90831f6d", "generation_seconds_batch": 16.544586537987925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7292227745056152, "target_z": 1.4609375869576269, "ce": 6.300607204437256, "genuineness": 4.6203155517578125, "identity": 0.7087670564651489, "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_AUTH_reg/befdb12e2a52ab98b364fbcfe1cdeab2205daed2dbfe801fbdb42da39d09ba1e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "bf1ec742e781038c28a628baa25dbb5a5d30d2e99d305b9caa94518567f75ca1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "210a69feeef0e6f8fc0a075c811ec7ba500f00eb73e09817dc6fd55836d02709", "generation_seconds_batch": 27.879800538066775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.205295085906982, "genuineness": 4.193545341491699, "identity": 0.7889495491981506, "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_AUTH_reg/bf1ec742e781038c28a628baa25dbb5a5d30d2e99d305b9caa94518567f75ca1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "bf77d2556c9882c100206fe5a641fedddfd2eaf1e52347d25314628c373bd898", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4c612836ba11ae84472022fdd22961c66a87d8ba045663f84a03d9b451c71047", "generation_seconds_batch": 24.976513386005536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.1894097328186035, "genuineness": 4.137060642242432, "identity": 0.5663702487945557, "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_AUTH_reg/bf77d2556c9882c100206fe5a641fedddfd2eaf1e52347d25314628c373bd898.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "bfdf065f63d8edfae1ee8b90b06bf257a234b9cff4eec5823887c588f5485ecd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f80d8e084dc104de5706e21b659e4decd7cb9c242680b37e02a943892d0a3f27", "generation_seconds_batch": 19.611200220067985, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.575014352798462, "target_z": 1.312499992828237, "ce": 5.903473377227783, "genuineness": 4.356721878051758, "identity": 0.7044963836669922, "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_AUTH_reg/bfdf065f63d8edfae1ee8b90b06bf257a234b9cff4eec5823887c588f5485ecd.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": "c030bc67489782082978b422f93acb3fa2ca8d1b9ea9f37bfb05763a9bed9254", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a7edd4270bc73e9f457df089544e27d4256c497e2abaca1e46aa881ec218c661", "generation_seconds_batch": 18.016439547063783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.435632705688477, "genuineness": 3.898571252822876, "identity": 0.22927968204021454, "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_AUTH_reg/c030bc67489782082978b422f93acb3fa2ca8d1b9ea9f37bfb05763a9bed9254.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "c06d40111a0a35b3ebb14dc9e0f40ad1c9259f1416b831548baa10b3e70772dd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f2727db6c047c4ee0d54186ffca8eaa7e98fe4e958456ad430c7ded974a76c3", "generation_seconds_batch": 28.117583174956962, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9483609199523926, "target_z": 1.671875087854097, "ce": 6.38003396987915, "genuineness": 4.262581825256348, "identity": 0.3910911977291107, "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_AUTH_reg/c06d40111a0a35b3ebb14dc9e0f40ad1c9259f1416b831548baa10b3e70772dd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "c19ad7fa57924f3aca11e2b17c13c6e158ed6dd0c66a252f1d8d35d585fbde07", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01cf485d06a6482992f5c620f9785bec980a755b9aa8d9f526a798871a136cc1", "generation_seconds_batch": 25.66594668000471, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.4912309646606445, "genuineness": 4.068024158477783, "identity": 0.5513925552368164, "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_AUTH_reg/c19ad7fa57924f3aca11e2b17c13c6e158ed6dd0c66a252f1d8d35d585fbde07.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "c21778b50c62c0e067026e3108418984348f3565cdb1bd45488ed91af763d69b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7408a4cc8cdc9877a56b422522a8bb9d0bd438984605a23adb1f5fc775755964", "generation_seconds_batch": 18.66718391003087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.308549880981445, "genuineness": 4.042920112609863, "identity": 0.1321885734796524, "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_AUTH_reg/c21778b50c62c0e067026e3108418984348f3565cdb1bd45488ed91af763d69b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "c2fcea5a8994223cc5164b7c90eb4d94ee67284e0bf3f74dcc36fe911b31c783", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d30417bdcd7b340e63e12d0e20dc901af608cba9f3fec9249f24a94968332a7", "generation_seconds_batch": 25.60334236000199, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.435632705688477, "genuineness": 4.093128204345703, "identity": 0.524703323841095, "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_AUTH_reg/c2fcea5a8994223cc5164b7c90eb4d94ee67284e0bf3f74dcc36fe911b31c783.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "c306339d6778111684beed6d37b8b07f3386f9e9571fb1aa53554814a6eaa310", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b1c94cbf8892d25cdbe70dc252a90801a3008309c75cde7a81fe81279edc175", "generation_seconds_batch": 26.62451807805337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.062326908111572, "genuineness": 4.218649387359619, "identity": 0.6301605701446533, "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_AUTH_reg/c306339d6778111684beed6d37b8b07f3386f9e9571fb1aa53554814a6eaa310.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "c33d15af0606d01522d4117a2525665602361b9a1dfcad30026d228ca046fb58", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52aedcd47d785c02859fc1e85e72b308062b70a118f2ca9c7d005560b30485bc", "generation_seconds_batch": 22.591598783037625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.05963397026062, "target_z": 0.8164063087188098, "ce": 6.300607204437256, "genuineness": 4.2751336097717285, "identity": 0.40801966190338135, "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_AUTH_reg/c33d15af0606d01522d4117a2525665602361b9a1dfcad30026d228ca046fb58.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "f91a47854ced797d7224dd05339f0ba35b144c40e788ff97e6b2ad33f615f7ec", "generation_seconds_batch": 18.08032832108438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.1894097328186035, "genuineness": 4.18099308013916, "identity": 0.6403303146362305, "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_AUTH_reg/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "c3b118e671b4be6ac257dbcd97ea87c937c8ab89cd19eda8477e96ce38741855", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b00192d5bb4ac1ec5b1a8bcc4d259235a6e361a643d5c259a26ab07a2205df6", "generation_seconds_batch": 25.83499683998525, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.348263263702393, "genuineness": 3.8044304847717285, "identity": 0.45983412861824036, "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_AUTH_reg/c3b118e671b4be6ac257dbcd97ea87c937c8ab89cd19eda8477e96ce38741855.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "c48a2d4679984a5f282b8aee1e2aec1457dca457f186af812d685338e95a2bfa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46d428b7ed94a55ab439783b7e9de120f58e16b7fc85e8881de22cb6da9df958", "generation_seconds_batch": 19.502909668022767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.229123115539551, "genuineness": 4.149612903594971, "identity": 0.2077711522579193, "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_AUTH_reg/c48a2d4679984a5f282b8aee1e2aec1457dca457f186af812d685338e95a2bfa.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "90d100727c08fed98ac6dd6fe9e39275df9ec819d2491cbc910544ba5dff91dc", "generation_seconds_batch": 19.145839125965722, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.6497901678085327, "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_AUTH_reg/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "c4f6d60e1f1104fc4c429d3fc1eee6789f4144760258225db99679f92cf619b3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a74835e5a2022e736f7759d604dd148240b79ce7d49214fd8f875ae50677e30", "generation_seconds_batch": 26.72619521606248, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.332377910614014, "genuineness": 4.13078498840332, "identity": 0.39270156621932983, "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_AUTH_reg/c4f6d60e1f1104fc4c429d3fc1eee6789f4144760258225db99679f92cf619b3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "c51fc2550a990018b7fb80a9e288e4174471196985465d4d0cab847b7966b2aa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "326b6ec96c631a04045ee78bd2722b3dea7e9d05b45d7ea4d23268bbd1684494", "generation_seconds_batch": 25.361834149924107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.932128429412842, "target_z": 1.6562500537882228, "ce": 6.070269584655762, "genuineness": 4.557555198669434, "identity": 0.6885522603988647, "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_AUTH_reg/c51fc2550a990018b7fb80a9e288e4174471196985465d4d0cab847b7966b2aa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "c55572e6f17777b9f0c654ac20baa3e0b63d5dea12dd63eb39f7e54b51f46c91", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d69e5a40f6bbaac33f890e2ac6d350627e025aac0ea20e9a6bbf55d38337bbe", "generation_seconds_batch": 27.01790074305609, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.630124092102051, "target_z": 2.32812514164232, "ce": 5.796247482299805, "genuineness": 4.068024158477783, "identity": 0.6413789987564087, "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_AUTH_reg/c55572e6f17777b9f0c654ac20baa3e0b63d5dea12dd63eb39f7e54b51f46c91.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "c578e1f8f09148316ab3479475e9c1d168edf9ab3ca2692f08137947a986f430", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b91279dfbc6f90ad08d2067d42527f83c2f10cffd6eee6164afce69330af91d", "generation_seconds_batch": 26.178430213942192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2747139930725098, "target_z": 1.0234375510988116, "ce": 6.252951145172119, "genuineness": 4.7081804275512695, "identity": 0.7315940856933594, "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_AUTH_reg/c578e1f8f09148316ab3479475e9c1d168edf9ab3ca2692f08137947a986f430.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "c5886720d056d1691aa650c6599adea9734876b94294335f09d0c67480755213", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "12928a3973985efee050c7c73a1a29e459dc708e3a5bd35c2861c5137944f0a6", "generation_seconds_batch": 19.932358084013686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 5.919358730316162, "genuineness": 4.193545341491699, "identity": 0.7449928522109985, "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_AUTH_reg/c5886720d056d1691aa650c6599adea9734876b94294335f09d0c67480755213.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "c644fe4664442b59cc76d742f07bc7517dc71d378e6f00df3a1247c0aadf03fe", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5e5dfb70584cd16a493d957e4007e598b03bd118739cd969fe8e98ab6e852d2a", "generation_seconds_batch": 26.98474163992796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5668981075286865, "target_z": 1.3046874757952998, "ce": 6.133810997009277, "genuineness": 4.3881025314331055, "identity": 0.5154921412467957, "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_AUTH_reg/c644fe4664442b59cc76d742f07bc7517dc71d378e6f00df3a1247c0aadf03fe.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "c64fbb2770956fab16f462337072d67f3e3295d5feb6557dd72443221976754d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe40c6ca27857350768376e97beb8d27fd3918f86fa18954fe2aec3eb2b7d0b0", "generation_seconds_batch": 25.13241223699879, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.279422283172607, "target_z": 2.9531251272987937, "ce": 6.141753673553467, "genuineness": 3.9927117824554443, "identity": 0.6276448965072632, "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_AUTH_reg/c64fbb2770956fab16f462337072d67f3e3295d5feb6557dd72443221976754d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "c6666d76527fda018c21607a9586e897d145a941e8b9c09fcf5a26129a48c72c", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c2bc3921945de4f038f3b814f00b966bb1864f8a746facea2e99a0d741359dc5", "generation_seconds_batch": 19.060482084052637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.299062728881836, "target_z": 1.0468751021976233, "ce": 6.1894097328186035, "genuineness": 4.262581825256348, "identity": 0.4174115061759949, "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_AUTH_reg/c6666d76527fda018c21607a9586e897d145a941e8b9c09fcf5a26129a48c72c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "c6e4887fb3d06aaa68d81a5df19310e651f2e6ddbf4294ab0911cb62579cb8b4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "110739a411b73049e1752d9502ef11cdb9ff236ffdebe93f12da9b02d60640b3", "generation_seconds_batch": 22.591598783037625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4208059310913086, "target_z": 1.1640623986988472, "ce": 6.38797664642334, "genuineness": 4.381826400756836, "identity": 0.3880002796649933, "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_AUTH_reg/c6e4887fb3d06aaa68d81a5df19310e651f2e6ddbf4294ab0911cb62579cb8b4.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "c71ddfc735114bf46ca4393e8cfa4f390f03c677b63aef519d34c907f28d2ae8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f9cfad0c74023fb8e5dd704a6f05370b8776ae9741a5a55ce29c3e997bd348f0", "generation_seconds_batch": 18.92815965693444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.678821086883545, "target_z": 2.374999784847109, "ce": 6.459460258483887, "genuineness": 4.218649387359619, "identity": 0.5939127206802368, "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_AUTH_reg/c71ddfc735114bf46ca4393e8cfa4f390f03c677b63aef519d34c907f28d2ae8.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": "c7d2a72e5f084e4a7b73fb7aa812a5102e4d8e3a585473437d18019508ba304b", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd83359ae502b82d87ab55b9f331505a36b5d3d3c0b8269cc3b22b5ee1c60f26", "generation_seconds_batch": 16.673505916027352, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 6.205295085906982, "genuineness": 4.2249250411987305, "identity": 0.6458717584609985, "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_AUTH_reg/c7d2a72e5f084e4a7b73fb7aa812a5102e4d8e3a585473437d18019508ba304b.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "31cda4aee98a0d377325d67677003961c2a6c8f64b0da498856d899b438d8ad0", "generation_seconds_batch": 18.61614672304131, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8022687435150146, "target_z": 1.5312500107576446, "ce": 6.09409761428833, "genuineness": 4.218649387359619, "identity": 0.6659699082374573, "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_AUTH_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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "703b07c239b34595d5cc2da371a787df5e6ffd7df06f103a9d4de06f75ab28b6", "generation_seconds_batch": 19.482396466075443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9500648975372314, "target_z": 0.7109375582705747, "ce": 6.459460258483887, "genuineness": 4.432034492492676, "identity": 0.5383585691452026, "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_AUTH_reg/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "c9eafec78d670fb118612248a05bfe7a781fe3b9423ed5c516cf903422914181", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "455035d031e5d662a37b377178164c18f3c7c80719c060dfdc70866e31a49261", "generation_seconds_batch": 24.71703683992382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.181467056274414, "genuineness": 4.074300289154053, "identity": 0.5231915712356567, "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_AUTH_reg/c9eafec78d670fb118612248a05bfe7a781fe3b9423ed5c516cf903422914181.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "ca1422c241f5b7c2aea05fb92cc828ceee3925f77761faa5c01e7f75f3e4d068", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1df4e34677e67a0cb0690d5134439a9d3ffd8649c935a44da16414003eb38239", "generation_seconds_batch": 27.60471205005888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.915895938873291, "target_z": 1.6406250197223484, "ce": 6.1894097328186035, "genuineness": 4.381826400756836, "identity": 0.4399706721305847, "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_AUTH_reg/ca1422c241f5b7c2aea05fb92cc828ceee3925f77761faa5c01e7f75f3e4d068.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "ca6c4cc67087f0b9235c0b7cea07cac07cbc24cdcd3082b5cfd5d3c77a2691e0", "bucket": "vn_S_AUTH_reg", "phase": "wording", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": "952b667177eddc11b1e6ce59231849abe952d04d0c4e15864d48c6d12ab8f06a", "generation_seconds_batch": 19.012173319002613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.956477165222168, "target_z": 1.6796876048870344, "ce": 6.340320587158203, "genuineness": 4.557555198669434, "identity": 0.5232607126235962, "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_AUTH_reg/ca6c4cc67087f0b9235c0b7cea07cac07cbc24cdcd3082b5cfd5d3c77a2691e0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "cba5db931e03bef6424cd5fba122a035068843e1b53053263e14523107a43451", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3053a9b54fc47a75ca9a3e99c1daa18a5120d9396d39f481b27f24c4be38d978", "generation_seconds_batch": 26.380035863956437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.1020402908325195, "genuineness": 4.243753433227539, "identity": 0.4292369782924652, "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_AUTH_reg/cba5db931e03bef6424cd5fba122a035068843e1b53053263e14523107a43451.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "cc46508b21387eda3782799d873e508dd12c0c5832c6321d508b42d86a6baec1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "604386a83d698cb6b4d449bfb76f2bb578ce8043c0303638bed76f9b57da0e94", "generation_seconds_batch": 25.589322187006474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.165581703186035, "genuineness": 4.394378662109375, "identity": 0.8215407729148865, "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_AUTH_reg/cc46508b21387eda3782799d873e508dd12c0c5832c6321d508b42d86a6baec1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "cc5509f13eab882c65ff634a6c49482bd65ad268690b500820844869bf0f3e65", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73ff69f87f09b1de8ca18c3366229e2510fb3a2e9bfcdad92842212dc224f595", "generation_seconds_batch": 25.257621766999364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.205295085906982, "genuineness": 4.105680465698242, "identity": 0.22233763337135315, "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_AUTH_reg/cc5509f13eab882c65ff634a6c49482bd65ad268690b500820844869bf0f3e65.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "cc9b5ec52f33f9e85782c84fb602b37d75b4850ee4c75f8a45e8659dfbf5e6f8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bb593d4f6fa4de100986046ecb1b2dfe811edb9e3bae789829aeaace1a2df5c", "generation_seconds_batch": 26.32413364900276, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.230074644088745, "target_z": 0.980468707417657, "ce": 6.419747352600098, "genuineness": 4.319066047668457, "identity": 0.43401801586151123, "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_AUTH_reg/cc9b5ec52f33f9e85782c84fb602b37d75b4850ee4c75f8a45e8659dfbf5e6f8.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": "cce9101b75be620114d93d69b2ec9ce4b5c7ec889b06e3c2a269e2be8672a382", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8363ab265f2204de45d657d0b4a9da6d9feec98ffb01c4f45cc706087ee7544d", "generation_seconds_batch": 25.93472476908937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.205295085906982, "genuineness": 3.986435890197754, "identity": 0.7801116704940796, "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_AUTH_reg/cce9101b75be620114d93d69b2ec9ce4b5c7ec889b06e3c2a269e2be8672a382.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": "cd112f9754d4c2d2d4344513782229aaa1b55fc6dd98462f66c7dff4d4ad3671", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23e1040a63d2a6c292b8d065f1655ea991fbbf1dafc782aec1e2e5b4ff7baf38", "generation_seconds_batch": 22.26621481194161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9223079681396484, "target_z": 2.609374836842391, "ce": 6.395919322967529, "genuineness": 4.243753433227539, "identity": 0.5791654586791992, "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_AUTH_reg/cd112f9754d4c2d2d4344513782229aaa1b55fc6dd98462f66c7dff4d4ad3671.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": "cd12e30738eadf74f8ee4153be1c5704a4b856d520474ee30dec25bd3cb03858", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efbd8fc3f521f66d87b2dd8da3d564128a2de7df5c046f8f3ce720a7ede12dd2", "generation_seconds_batch": 22.130793037940748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.630124092102051, "target_z": 2.32812514164232, "ce": 6.316492557525635, "genuineness": 3.898571252822876, "identity": 0.6430753469467163, "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_AUTH_reg/cd12e30738eadf74f8ee4153be1c5704a4b856d520474ee30dec25bd3cb03858.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": "cd345695e6acfc3a4ac87400dd02b3edf27275268d4dffee48784bbc55329524", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b2dcee682b8e95f2ff828fba84185244a3513405492859664ff31aac8bb6ef38", "generation_seconds_batch": 25.588872020016424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06198787689209, "target_z": 1.7812498673223838, "ce": 6.149696350097656, "genuineness": 4.306513786315918, "identity": 0.5897564888000488, "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_AUTH_reg/cd345695e6acfc3a4ac87400dd02b3edf27275268d4dffee48784bbc55329524.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "cd54f31d9609aca4a2e055306afd0af91228f342dc55c879f088f2d97099eafa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0e2cc2342c4406fdcde3748d93c967d22af1cbce33497ee45d5be1212a0446b", "generation_seconds_batch": 26.915509214042686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0376393795013428, "target_z": 1.7578125457199893, "ce": 6.062326908111572, "genuineness": 4.432034492492676, "identity": 0.6887702941894531, "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_AUTH_reg/cd54f31d9609aca4a2e055306afd0af91228f342dc55c879f088f2d97099eafa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "ce163420fd1b1785154f30d02dda69d03493dc3c5554636879dba9ca5f7f31c5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90ba594da64ef6db40f0387e31d55fddc19a202f1049ba9ff5e84ddf70bac49f", "generation_seconds_batch": 25.89979123591911, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.41180419921875, "genuineness": 4.432034492492676, "identity": 0.5509438514709473, "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_AUTH_reg/ce163420fd1b1785154f30d02dda69d03493dc3c5554636879dba9ca5f7f31c5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "ce2fd5268b4c63294372e6013c5dbcb3e4bc52f738686119ce2a738aca56fd7a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "574f13e6df1e4a1dfe5b7fe49f5c68e4c25b6b8efbc823ad36a599c4b4c869f0", "generation_seconds_batch": 28.278084215940908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.070269584655762, "genuineness": 3.9048471450805664, "identity": 0.4304637610912323, "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_AUTH_reg/ce2fd5268b4c63294372e6013c5dbcb3e4bc52f738686119ce2a738aca56fd7a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "ce680b198d8cbdd425ade75f781f22c7c928e6029fd8b178f43f96d9881f3192", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31f7ec342b6be6df17a62a141cabc3165db0e9d916550ce21e787f216dc86090", "generation_seconds_batch": 26.34820273704827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.292664527893066, "genuineness": 4.256305694580078, "identity": 0.5151458382606506, "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_AUTH_reg/ce680b198d8cbdd425ade75f781f22c7c928e6029fd8b178f43f96d9881f3192.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "ce96672a8cda333e2f1e8e8b86286835e458f04489da1a4953a0fa59e6196811", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "021071e96ca267540cd873f253f56b23ffd255a7dcedf5e79e7ccd220d778050", "generation_seconds_batch": 28.29972495196853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.36399245262146, "target_z": 1.1093750089647039, "ce": 6.24500846862793, "genuineness": 4.595211505889893, "identity": 0.529657781124115, "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_AUTH_reg/ce96672a8cda333e2f1e8e8b86286835e458f04489da1a4953a0fa59e6196811.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "ceb84bfc453956cea4c16925bf6e5c31e101e1bb91f9cb8f94a569e12270ae52", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b71424dfe3b23066db0bd9746bb1c58f0fce108e0d6cc141870957144f6e61f4", "generation_seconds_batch": 25.710229006945156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 6.181467056274414, "genuineness": 3.8420867919921875, "identity": 0.13263359665870667, "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_AUTH_reg/ceb84bfc453956cea4c16925bf6e5c31e101e1bb91f9cb8f94a569e12270ae52.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "cebf9cdfb0edc270ce7b383d48e29929dcc111cf73f5c3442568084cfd08e8a4", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "576cd00bd75326ce09dcfd3866477209ce0af262e5cbbd128e04cb5d8e06cb20", "generation_seconds_batch": 24.62843315396458, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.062326908111572, "genuineness": 4.036643981933594, "identity": 0.5692042708396912, "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_AUTH_reg/cebf9cdfb0edc270ce7b383d48e29929dcc111cf73f5c3442568084cfd08e8a4.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": "cec1dd119767135e98ce5442665bfe099d053e50bb00863579da244a9eec06fa", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b964255a8500395683bb289f74a0270211863dd39f0f51b90745fe6e7d1f2104", "generation_seconds_batch": 22.072075044037774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13332986831665, "target_z": 2.812499820705924, "ce": 6.24500846862793, "genuineness": 3.8546390533447266, "identity": 0.5607967376708984, "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_AUTH_reg/cec1dd119767135e98ce5442665bfe099d053e50bb00863579da244a9eec06fa.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "cecf5eb081b9b9a8e2cbdbddeb75376f3e55939e5daf0c331ee5ed996e46e7fb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d45c3ef9e1fce6d355c5f119e4c1860548ccfebda17b4e6d9f9ca2fc7b581e19", "generation_seconds_batch": 26.27278962091077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 5.974957466125488, "genuineness": 4.099404335021973, "identity": 0.5678720474243164, "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_AUTH_reg/cecf5eb081b9b9a8e2cbdbddeb75376f3e55939e5daf0c331ee5ed996e46e7fb.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "04fdf1aee2d3316a50dc325fd45d909a1699668edab91c344520f73708e56dc5", "generation_seconds_batch": 19.897955678985454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.284721851348877, "genuineness": 4.168440818786621, "identity": 0.6595361828804016, "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_AUTH_reg/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.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": "cfae203a224cb16851e35fccc9bab88fc080582ef953fa7e67df6a53990ab181", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f99959273aeebf2b268a1463eaf8c07facf56b7e89d229f38b42459918204b2b", "generation_seconds_batch": 22.47037289501168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.419747352600098, "genuineness": 3.9550557136535645, "identity": 0.3018414378166199, "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_AUTH_reg/cfae203a224cb16851e35fccc9bab88fc080582ef953fa7e67df6a53990ab181.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "cfdfec3258391ffd28e87aab2ff58add9a1d6c0b20bb16e18765f843df633ff7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ba27e9a1646e62b97aab36924e963a8a775b129104976d9966ba9fb409490a3b", "generation_seconds_batch": 27.428042290033773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7779200077056885, "target_z": 1.507812459658833, "ce": 6.340320587158203, "genuineness": 4.444586753845215, "identity": 0.5309211611747742, "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_AUTH_reg/cfdfec3258391ffd28e87aab2ff58add9a1d6c0b20bb16e18765f843df633ff7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "d0269d4a8106aeddc713ea500aa687a3a3cf935f923e8c4d369a01fdc759636b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "441d214eb7c555ee92176e2775a15914bad9b87f5ee4c8d44767c562481426a2", "generation_seconds_batch": 25.407685216050595, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.045755386352539, "target_z": 1.7656248332565094, "ce": 5.95112943649292, "genuineness": 4.080576419830322, "identity": 0.6975163221359253, "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_AUTH_reg/d0269d4a8106aeddc713ea500aa687a3a3cf935f923e8c4d369a01fdc759636b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "d06f4f9c28fdd7e8675f58bbe6a6b4718ca4436d80bc6b6557b30c60189f62cb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d52dd4642e7fb87c7a9098d0d7702d85b9e6a9c6b1565e1d9e67e2fbe213aa0", "generation_seconds_batch": 24.332862335955724, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8185012340545654, "target_z": 1.546875044823519, "ce": 6.022613525390625, "genuineness": 4.482243061065674, "identity": 0.5391498804092407, "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_AUTH_reg/d06f4f9c28fdd7e8675f58bbe6a6b4718ca4436d80bc6b6557b30c60189f62cb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "d098604c2b552888031915943a848cf7515eb0e8befa3d524ffbea4fa30002a7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d987ceb5b09acb174ba4264c92178e652a7f619ffc65b7f878690a246405d82e", "generation_seconds_batch": 25.204119879985228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8202052116394043, "target_z": 0.5859375152399965, "ce": 6.006728172302246, "genuineness": 4.607763767242432, "identity": 0.7649682760238647, "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_AUTH_reg/d098604c2b552888031915943a848cf7515eb0e8befa3d524ffbea4fa30002a7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "d0c8f4e4e7119187be53a67c450fec009e6c88470fae1a5412e832a98c8c250d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bb5895c1ec2db33d49f06f135ad7dd1379c393c40ff6b79e859172c46709e2a", "generation_seconds_batch": 24.578243616968393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 5.927301406860352, "genuineness": 4.017816066741943, "identity": 0.5561618804931641, "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_AUTH_reg/d0c8f4e4e7119187be53a67c450fec009e6c88470fae1a5412e832a98c8c250d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "d108b8026587905e254008acd13fcd24a320a2f0951be4c2933c8c24d841ab9c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9f61987e4b4d111b35426131484b8f7dedbb3ba1e985de923b40722db859310", "generation_seconds_batch": 19.236081793089397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06198787689209, "target_z": 1.7812498673223838, "ce": 6.340320587158203, "genuineness": 4.281409740447998, "identity": 0.21915867924690247, "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_AUTH_reg/d108b8026587905e254008acd13fcd24a320a2f0951be4c2933c8c24d841ab9c.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "c0665f2c038bfba3d4238dbf553ebde61ad6f813f06702c166d7ef808db54206", "generation_seconds_batch": 17.923304528929293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.704874038696289, "target_z": 1.4375000358588153, "ce": 6.332377910614014, "genuineness": 4.124508857727051, "identity": 0.23474356532096863, "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_AUTH_reg/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.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": "d1a482bf47776918baf37fa3566253643b36acf351ef37af120f5019e6025a0d", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc5844d11cff76c1f0d25b3eb77f29b6b201cab887576fae3641c9e536e4ff35", "generation_seconds_batch": 16.61147071502637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393049240112305, "target_z": 3.0624999067670804, "ce": 6.165581703186035, "genuineness": 4.080576419830322, "identity": 0.20634549856185913, "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_AUTH_reg/d1a482bf47776918baf37fa3566253643b36acf351ef37af120f5019e6025a0d.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "3b584e2c2bf8685cede594c801ecce4c41f9868f50a3b8e9a9bf03f896ce4d14", "generation_seconds_batch": 15.72046716301702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2405447959899902, "target_z": 1.9531247830541683, "ce": 6.1894097328186035, "genuineness": 4.061748504638672, "identity": 0.6969413757324219, "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_AUTH_reg/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "d329e04d8802c0d9d8b11f76e58f1db718904b63a2dc57af5ea04a920bbd923d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b42ee73c2b9a03032392a57e4456484e300699f9e3caa0ab29a95c12342c4bff", "generation_seconds_batch": 24.647520172060467, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.292664527893066, "genuineness": 4.306513786315918, "identity": 0.32795092463493347, "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_AUTH_reg/d329e04d8802c0d9d8b11f76e58f1db718904b63a2dc57af5ea04a920bbd923d.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "a21aa9f05171e72350136d818ac0640d96683a17008da74d75d0577fd27fcdc3", "generation_seconds_batch": 18.076764025026932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 5.879645347595215, "genuineness": 4.532451152801514, "identity": 0.6872132420539856, "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_AUTH_reg/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "d42574c3964cc6f4f9de09864bd6d32632041fcf7f45e1a328994b34c08aaa22", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06fee6bf24281aacffcf416ad7ab0ae954999360af1d2acd606fc1b434393a94", "generation_seconds_batch": 19.040464148973115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.268836498260498, "genuineness": 4.15588903427124, "identity": 0.35975852608680725, "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_AUTH_reg/d42574c3964cc6f4f9de09864bd6d32632041fcf7f45e1a328994b34c08aaa22.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "d4d3c5f5e376ccddb8a6ffc5466ea8d821fc732415d751ba2dd68e515d504136", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64f96534ec641fcc89449dff809ccc615091ca0a76e55c99530f5cb02f7ff87c", "generation_seconds_batch": 25.13642307196278, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3315277099609375, "target_z": 1.0781251703293722, "ce": 6.395919322967529, "genuineness": 4.250029563903809, "identity": 0.5024032592773438, "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_AUTH_reg/d4d3c5f5e376ccddb8a6ffc5466ea8d821fc732415d751ba2dd68e515d504136.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "d4d48c35ab667bfd7794b3f24e6bd959e020a19ef810f7a7c392a310d78ff0c2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eddceda01bca29f95f2737d2cacca06565668a380d8efae896556fe385c5cc91", "generation_seconds_batch": 26.62451807805337, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.0684003829956055, "target_z": 2.7500001434352606, "ce": 6.205295085906982, "genuineness": 3.9173994064331055, "identity": 0.6413432359695435, "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_AUTH_reg/d4d48c35ab667bfd7794b3f24e6bd959e020a19ef810f7a7c392a310d78ff0c2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "d5a7df9660a74d53c8703d78c5ad9fcd23b18cc55fb2a2bd98067bc359e131ae", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb32988f6d6e82cd461af837c20e319860bc320f138bd9e5ad8c676a1c137356", "generation_seconds_batch": 25.78359319700394, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.824913501739502, "target_z": 2.5156250914399787, "ce": 5.903473377227783, "genuineness": 3.9550557136535645, "identity": 0.5664103031158447, "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_AUTH_reg/d5a7df9660a74d53c8703d78c5ad9fcd23b18cc55fb2a2bd98067bc359e131ae.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "d5bcaff69bd2780f4e4747602a00bedd81adef9648e7f86e82274415001559f0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15da6641853860e9aeb2ae78fa42d12f9bb85bef61879776f788931edcec39a2", "generation_seconds_batch": 19.221954049076885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0109364986419678, "target_z": 0.7695312065211866, "ce": 6.41180419921875, "genuineness": 4.482243061065674, "identity": 0.3097951114177704, "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_AUTH_reg/d5bcaff69bd2780f4e4747602a00bedd81adef9648e7f86e82274415001559f0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "d5d218c5d8ef184d269a51118f2596feaeffd3a709024b39e137cbe6047c8897", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ece071b53414613f6c1ae76bb35ba09f99d0325dcb202c263fc5e02c77e55702", "generation_seconds_batch": 27.127703204983845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7941524982452393, "target_z": 1.5234374937247073, "ce": 6.229123115539551, "genuineness": 4.20609712600708, "identity": 0.5737529993057251, "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_AUTH_reg/d5d218c5d8ef184d269a51118f2596feaeffd3a709024b39e137cbe6047c8897.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "d6087dcbcd5dc949247cb808929aba77b25eddc286fa8d62e396257ef064f1b5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf427085e237112b5aba7c9eea38f7eb9d599e82d6c9c5b9612a90fcb0d25c98", "generation_seconds_batch": 28.110335026984103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.850965976715088, "target_z": 1.5781248834588506, "ce": 6.109982967376709, "genuineness": 4.319066047668457, "identity": 0.6266316175460815, "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_AUTH_reg/d6087dcbcd5dc949247cb808929aba77b25eddc286fa8d62e396257ef064f1b5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "d637e1cc8fbef3ea21b7b6ba9a2ef658552d33ee3c71700bd6aa411b742a80d2", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ff487b764c189610267d20db6b9ddee7cab4c8a6ad951066cd49fcb9b3c86f8", "generation_seconds_batch": 27.719575587077998, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.324435234069824, "genuineness": 4.080576419830322, "identity": 0.426643967628479, "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_AUTH_reg/d637e1cc8fbef3ea21b7b6ba9a2ef658552d33ee3c71700bd6aa411b742a80d2.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "d65c608cdb53e576f73a029713f4cf46b6703c4af45c60cb5883033b5bc50fa0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f96e42614b37754fc965361924a280bfe64d805fe7a6c7a3cd9c5d017ca6c7fa", "generation_seconds_batch": 25.233970609027892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2405447959899902, "target_z": 1.9531247830541683, "ce": 6.038498878479004, "genuineness": 4.15588903427124, "identity": 0.5358759760856628, "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_AUTH_reg/d65c608cdb53e576f73a029713f4cf46b6703c4af45c60cb5883033b5bc50fa0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "d6863a5b3425c1e46ce56d296a6d1df5a8dacbd66d9751bb89215a49d40f8125", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4415b1192d84654d2043d4781e1c5a218f47a5ea39371ca9d001808165240e51", "generation_seconds_batch": 25.52621907205321, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9987621307373047, "target_z": 0.7578124309717807, "ce": 6.507116317749023, "genuineness": 4.362998008728027, "identity": 0.32920098304748535, "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_AUTH_reg/d6863a5b3425c1e46ce56d296a6d1df5a8dacbd66d9751bb89215a49d40f8125.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "d6d0383b296ea692da533313277226963abfba241eed880e02bfc1df07250d99", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48967e168999e9197d4260810de50ad0253690e1f0f0e8f6e5c3b0b307264dde", "generation_seconds_batch": 26.843750506988727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.467799186706543, "target_z": 2.1718748009835758, "ce": 6.24500846862793, "genuineness": 3.8169827461242676, "identity": 0.16360436379909515, "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_AUTH_reg/d6d0383b296ea692da533313277226963abfba241eed880e02bfc1df07250d99.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "d721a6fcdcc05b42a08d8e2fb617c38b43e52a96cf197f5e23b97a62623492cb", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7360c7593c87b6034da44558fd82c8dd6da71cbe9af1778bde6855bfb1c29c3", "generation_seconds_batch": 20.28801698004827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393049240112305, "target_z": 3.0624999067670804, "ce": 6.221180438995361, "genuineness": 4.168440818786621, "identity": 0.5176730155944824, "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_AUTH_reg/d721a6fcdcc05b42a08d8e2fb617c38b43e52a96cf197f5e23b97a62623492cb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "d721ddfd1bb75f947d6262cce5e48fed188d1e6fb2349b981b2918bb99ea4508", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54df4af66f76a79806ad20a24d914002d1c986d94df8b6152741b8f5c7fa0a4a", "generation_seconds_batch": 26.49455736298114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.3641486167907715, "genuineness": 4.394378662109375, "identity": 0.5545501708984375, "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_AUTH_reg/d721ddfd1bb75f947d6262cce5e48fed188d1e6fb2349b981b2918bb99ea4508.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": "d8d5fd4e407f9727f7849ef505a69594841d631f09c812ac7cca9cda6a40dadb", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e526029a0509aef0dad131c2cff9f2e6b898086c4d0c9878424d9208ed12d267", "generation_seconds_batch": 16.615672801970504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2486610412597656, "target_z": 1.9609373000871053, "ce": 6.078212261199951, "genuineness": 4.011539936065674, "identity": 0.6601564884185791, "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_AUTH_reg/d8d5fd4e407f9727f7849ef505a69594841d631f09c812ac7cca9cda6a40dadb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "d9b0f2dbf88f75212e3f51290305c91db6d929604f4f80ac446e49fd2015ebf6", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4db7622387870fd4d4bccedebf700ec24ef8202355931cdf566a4de08fe0ce93", "generation_seconds_batch": 24.256153885042295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9581811428070068, "target_z": 0.7187500753035119, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.7553088665008545, "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_AUTH_reg/d9b0f2dbf88f75212e3f51290305c91db6d929604f4f80ac446e49fd2015ebf6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "d9fc403337db987dccbb5742e6ce7db12f72dde0527a66f3a79ab0d6720faea7", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d1f4a070c85c7205533fb0599116aeea0495ab369c672291d5c80d8d739d4db", "generation_seconds_batch": 26.936580862966366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 6.340320587158203, "genuineness": 4.595211505889893, "identity": 0.612686038017273, "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_AUTH_reg/d9fc403337db987dccbb5742e6ce7db12f72dde0527a66f3a79ab0d6720faea7.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": "da01c376931a862aa455a4c66d8f4460c97aa7d95c8558494cc7f604dd8b0aeb", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52822f1b4c73daa1b91aa6ed02651970770611b1778155f0fe66d5f56981d0bf", "generation_seconds_batch": 25.69512859499082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2486610412597656, "target_z": 1.9609373000871053, "ce": 5.879645347595215, "genuineness": 4.20609712600708, "identity": 0.7340472340583801, "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_AUTH_reg/da01c376931a862aa455a4c66d8f4460c97aa7d95c8558494cc7f604dd8b0aeb.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "dad8293aaa3796f399933d6ce0bf20400da0a05ab4deccb0a1387e7dc3d73759", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d99d116e73e1d03f0a9edaf716cf14c403d5ceeba0fc1953c0cc5ec3308600a6", "generation_seconds_batch": 25.48607203492429, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1188015937805176, "target_z": 1.8359374865529443, "ce": 6.149696350097656, "genuineness": 4.093128204345703, "identity": 0.5694721937179565, "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_AUTH_reg/dad8293aaa3796f399933d6ce0bf20400da0a05ab4deccb0a1387e7dc3d73759.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "3efa7ddce34452657ffad34700b58d10a78a44fa091c79a42d1f81fee7ad2db6", "generation_seconds_batch": 16.35831957194023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.980825901031494, "target_z": 1.703125155985846, "ce": 6.395919322967529, "genuineness": 4.18726921081543, "identity": 0.35554516315460205, "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_AUTH_reg/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "db65872dfa72dcf9857d08f88cf38a4a07b7713339c24a856b508f4334160f0c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79614b0d5bbd4b8a751cecc45ffb902ca495f98e35feed8a21d4d55052fc79b8", "generation_seconds_batch": 19.348872640985064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.1894097328186035, "genuineness": 4.218649387359619, "identity": 0.7009038329124451, "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_AUTH_reg/db65872dfa72dcf9857d08f88cf38a4a07b7713339c24a856b508f4334160f0c.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": "db9cdea66ef572b0f881c68b1ba22017fc64d908df75a19fc2c9068bc38fe169", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6da9dfdda5739addccfd24eba0844ef2c7b53fc1f2e6ec334b5924b83da71045", "generation_seconds_batch": 25.588872020016424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1593828201293945, "target_z": 1.8750000717176303, "ce": 6.205295085906982, "genuineness": 4.2374773025512695, "identity": 0.6464627385139465, "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_AUTH_reg/db9cdea66ef572b0f881c68b1ba22017fc64d908df75a19fc2c9068bc38fe169.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "dbeb5b082704c2cae76ad049dd6313c6d6298cf226bc2ec7460c18f5c9cebb8f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "56410a8c86a42b7ea7df2c384c3734230b53accd019bdb7ff03ffc415e45040b", "generation_seconds_batch": 24.794424962019548, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5912468433380127, "target_z": 1.3281250268941114, "ce": 6.324435234069824, "genuineness": 4.18099308013916, "identity": 0.33774739503860474, "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_AUTH_reg/dbeb5b082704c2cae76ad049dd6313c6d6298cf226bc2ec7460c18f5c9cebb8f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "dc32ad990373ec8de85f9ac7be725f820241dee2aca1bddaecd9223d7649bda3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54a1fd54a3183ba95cb13bed74e397145125e507b11da5c18b10044f9f83932d", "generation_seconds_batch": 27.43370363500435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9645934104919434, "target_z": 1.6875001219199717, "ce": 6.197352409362793, "genuineness": 4.017816066741943, "identity": 0.43704548478126526, "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_AUTH_reg/dc32ad990373ec8de85f9ac7be725f820241dee2aca1bddaecd9223d7649bda3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "dc58d058daf46b2cab58e9ae455b39d595804a227099320d361683f5c3c205b5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8a0da1bd5c1ad32f559a4caaab92f868ca738b0d61a4f90b9d4ce63e8163d4cf", "generation_seconds_batch": 24.931762705906294, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.09409761428833, "genuineness": 4.068024158477783, "identity": 0.5756633281707764, "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_AUTH_reg/dc58d058daf46b2cab58e9ae455b39d595804a227099320d361683f5c3c205b5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "dce7ea87c91360fd8365d900ae37a8cab9f1c439c432a8888017940a53a6a294", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "005cbe14c862e36f73010fcef6a303b26cf6eb35519f633e03524cb449c7b1d5", "generation_seconds_batch": 18.591826800024137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.5336964726448059, "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_AUTH_reg/dce7ea87c91360fd8365d900ae37a8cab9f1c439c432a8888017940a53a6a294.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "dd32951af361011b86db7c533763dd6520071a60249fa1296955ad9b983dfc01", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed4e08020e1b1d169dda08296bbc3ce6dd9480d59f3dd3f1dd895b45937e0147", "generation_seconds_batch": 26.363409465993755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.229123115539551, "genuineness": 4.143336772918701, "identity": 0.08472831547260284, "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_AUTH_reg/dd32951af361011b86db7c533763dd6520071a60249fa1296955ad9b983dfc01.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "ddd2f968bab9601a0b7df6d2283b6d389c7b3a5b7028b49ca3b79675fc283489", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bad28b90c6dc9d6d5833a8222563fb4752de982c15b1e80ebd24a1f3d89ff7d2", "generation_seconds_batch": 24.22340861300472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.38797664642334, "genuineness": 4.193545341491699, "identity": 0.4860994219779968, "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_AUTH_reg/ddd2f968bab9601a0b7df6d2283b6d389c7b3a5b7028b49ca3b79675fc283489.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": "de01a71df3c97ce8d1af77c04a9c4ea40d75aed630883c17154c0c987a436ce2", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9ab810046b78962cfe770f3d714ccc61fa53d41de8d0222319a552f01bd1c7a", "generation_seconds_batch": 25.93472476908937, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.213237762451172, "genuineness": 4.055472373962402, "identity": 0.7424602508544922, "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_AUTH_reg/de01a71df3c97ce8d1af77c04a9c4ea40d75aed630883c17154c0c987a436ce2.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": "de352157dadc4846caa7f491a4b54aee9a4ad6eb487509f2bf7c0cb7409bc91e", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04a3fe7e6a6818a9b198119a761795897f8d3c521641f3b56fce27de6a692687", "generation_seconds_batch": 22.15884544304572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.054384231567383, "genuineness": 4.293961524963379, "identity": 0.7798642516136169, "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_AUTH_reg/de352157dadc4846caa7f491a4b54aee9a4ad6eb487509f2bf7c0cb7409bc91e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "de45ea3f8a44f5d9ec39112d79481ba12f756603f71537ed4f5a2b41f2b1342a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a34505e3da4c7e530dcb68a0f673f09b006daf10472fa85d62a0fc5537784d42", "generation_seconds_batch": 26.041506521985866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.395919322967529, "genuineness": 4.017816066741943, "identity": 0.5042243599891663, "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_AUTH_reg/de45ea3f8a44f5d9ec39112d79481ba12f756603f71537ed4f5a2b41f2b1342a.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": "de5abfd6d19515c0667e1fbe38e5a5a63b64b6f174f7d6eb9c7fd0a6f51fdb30", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33da71038b239fa0ef4043a589270c7bd52a6bf20421a4b951f300ba3ae4ddda", "generation_seconds_batch": 22.26621481194161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.340320587158203, "genuineness": 4.137060642242432, "identity": 0.5471354722976685, "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_AUTH_reg/de5abfd6d19515c0667e1fbe38e5a5a63b64b6f174f7d6eb9c7fd0a6f51fdb30.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "decc539d5935369216bc12e8a437fb4c8f1c517ff60608aea71dd90e94745c2d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d1a8f1a1f9b9227a656bc59e1e6a898b79810b4496df8bb9d55d07ebfd99062", "generation_seconds_batch": 26.495877828099765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.054384231567383, "genuineness": 4.13078498840332, "identity": 0.7232433557510376, "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_AUTH_reg/decc539d5935369216bc12e8a437fb4c8f1c517ff60608aea71dd90e94745c2d.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": "df3252d9a47a17f7d47e6edf27bdcb095977c5351a89dc7c78f74beb56d69ff7", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6743996eaa01516f5209b4b6f84019cfef5fef2aa5aca8479c7fabc63b2617b", "generation_seconds_batch": 26.237925509922206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.311887264251709, "target_z": 2.984375195430543, "ce": 6.205295085906982, "genuineness": 3.9425034523010254, "identity": 0.5077018141746521, "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_AUTH_reg/df3252d9a47a17f7d47e6edf27bdcb095977c5351a89dc7c78f74beb56d69ff7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "dfb1719824e56f5ed7eca49b8daf22251723467d62d9d313acda54ec9d75c070", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc3c1db4ce18d3164ea2d0fead2e5913896cb894ea9dc7d1effe4692d77d9a84", "generation_seconds_batch": 24.179298240924254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1170973777771, "target_z": 2.79687478664005, "ce": 5.974957466125488, "genuineness": 4.049196243286133, "identity": 0.698292076587677, "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_AUTH_reg/dfb1719824e56f5ed7eca49b8daf22251723467d62d9d313acda54ec9d75c070.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "dff33f926b5749ed8fda4614215cf96442de229a924000652f56a752e60f6946", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44dd123fd2f39269910888457ae9e3e187840162cfb47c67f2ad296fd4d1c6e7", "generation_seconds_batch": 25.519448391045444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.157639026641846, "genuineness": 4.293961524963379, "identity": 0.7422676086425781, "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_AUTH_reg/dff33f926b5749ed8fda4614215cf96442de229a924000652f56a752e60f6946.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "e0dc5f3ea6aec955c49609e2cc4ba5d88b2aedc6da145e56032903d46ab14140", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b9ba1968d141fbf753c6170667d40e387c539f17568e5542b56757869536af5", "generation_seconds_batch": 24.65907000703737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2243127822875977, "target_z": 1.937500207981128, "ce": 6.268836498260498, "genuineness": 4.331617832183838, "identity": 0.5944381952285767, "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_AUTH_reg/e0dc5f3ea6aec955c49609e2cc4ba5d88b2aedc6da145e56032903d46ab14140.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "e13c8208a822015b33371cdea7f8132a7bbae3d79413beb9fe5691fcaea40799", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbf9fa47bef2e6f8d0e32a68209162b66ba6338ecf53c8f74279b86ad9fdaf81", "generation_seconds_batch": 17.965504886000417, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.587838649749756, "target_z": 3.2499998565647394, "ce": 6.23706579208374, "genuineness": 4.143336772918701, "identity": 0.5850840210914612, "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_AUTH_reg/e13c8208a822015b33371cdea7f8132a7bbae3d79413beb9fe5691fcaea40799.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "e21f15d770ed81469ac755f6e45fd663d7cef56c36ac937c1289c831be8d0111", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b7af4e528bf5a9f2c673154b4b3c2b012809abaf5c84372f7928e42379a2fd6", "generation_seconds_batch": 25.080744503997266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8736109733581543, "target_z": 2.562500193637602, "ce": 6.340320587158203, "genuineness": 3.8797430992126465, "identity": 0.36317044496536255, "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_AUTH_reg/e21f15d770ed81469ac755f6e45fd663d7cef56c36ac937c1289c831be8d0111.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "e2499773fc680f7eafe68a496898bfe032df9d2a54b99675750219270ca78475", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "228e9324c6d8f3bffd95f81d4784e183acddbe6d73d65ce19f037543fa6f997f", "generation_seconds_batch": 27.70692950708326, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.24500846862793, "genuineness": 4.3127899169921875, "identity": 0.8302639722824097, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble what it 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_AUTH_reg/e2499773fc680f7eafe68a496898bfe032df9d2a54b99675750219270ca78475.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "e2616da75b9304d2f34d353364574640332bb9612597a2b65ab236d908c23c26", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff25e1fbab6b9574540ce7581ebb4bdce6bc666abaf1a89472c4fd2d004f2b47", "generation_seconds_batch": 25.82873369101435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.038498878479004, "genuineness": 4.557555198669434, "identity": 0.6496025323867798, "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_AUTH_reg/e2616da75b9304d2f34d353364574640332bb9612597a2b65ab236d908c23c26.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "e2e7e464f6ed975b2ade403e58a4dd380387e1458a5e6a631fb0fb945ce3aa69", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdcf8f630cca23df795c2daa096dbd0eae2e763cb5426478aeb4aa98ab6b4fa3", "generation_seconds_batch": 25.89878777100239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.234132766723633, "target_z": 0.9843749659341255, "ce": 6.09409761428833, "genuineness": 4.545003414154053, "identity": 0.7026688456535339, "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_AUTH_reg/e2e7e464f6ed975b2ade403e58a4dd380387e1458a5e6a631fb0fb945ce3aa69.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "e2f6cde207d03bccd197acc4d0aaded2cfcc23909aee7bfdecdcf209ff3b8eb0", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43e0e8cb7665d49969ddc64b6c12a5ecbe45a4ddc7e37825b7e75a347ce6bb2a", "generation_seconds_batch": 25.234663205919787, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9889419078826904, "target_z": 1.710937443522366, "ce": 6.157639026641846, "genuineness": 4.137060642242432, "identity": 0.6756413578987122, "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_AUTH_reg/e2f6cde207d03bccd197acc4d0aaded2cfcc23909aee7bfdecdcf209ff3b8eb0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "e349c270dd00769732fc25bf0fd07cb300e5032fd92865d18c086fbd34a6aa2c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b31486832835e59b90bc5a8f1f68d525d4a99162d4e9fe6c6735e56b0673863", "generation_seconds_batch": 19.004660979961045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 5.871703147888184, "genuineness": 4.18099308013916, "identity": 0.7291795015335083, "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_AUTH_reg/e349c270dd00769732fc25bf0fd07cb300e5032fd92865d18c086fbd34a6aa2c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "e34c1a79f87d99e7e90f9779f173dccea6f61de1fe0f139b7da18d9a1c750bb3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbdd22e19a334c083d23228774c61e34049b3d1298a63b432e9dce8069aa8f1f", "generation_seconds_batch": 30.237586554023437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0132906436920166, "target_z": 1.7343749946211777, "ce": 6.268836498260498, "genuineness": 4.068024158477783, "identity": 0.215273916721344, "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_AUTH_reg/e34c1a79f87d99e7e90f9779f173dccea6f61de1fe0f139b7da18d9a1c750bb3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "e3c323f4f72d4e23272fec138311fd2f3f585b17e26c51648ddde3e5a6bb2955", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "836d7886bffc21608786bd76e47e657d5defe50c0333dc1fd0dfd51bca5bdbae", "generation_seconds_batch": 25.80701417394448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.078212261199951, "genuineness": 4.243753433227539, "identity": 0.5965243577957153, "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_AUTH_reg/e3c323f4f72d4e23272fec138311fd2f3f585b17e26c51648ddde3e5a6bb2955.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "e3df3b2f8e9eb8b30d5a8b30459b2c4dde79f587064577e1aa780cb114534186", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e768d2b9b59595fd80c946267f9fc38eaaf5e58fd8bfdfccc18293da535a13d5", "generation_seconds_batch": 24.002248284989037, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.672782838344574, "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_AUTH_reg/e3df3b2f8e9eb8b30d5a8b30459b2c4dde79f587064577e1aa780cb114534186.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "e55673dfbbb7c0ac5a76f501f23424bcf2298efd8eff183f68bdde6e1b165b68", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5994d56f5bafb45c6207e45a4590eb6a66399405109935f2ac486de1602d6b89", "generation_seconds_batch": 18.168829747010022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.745455265045166, "target_z": 1.4765626210235012, "ce": 6.125868320465088, "genuineness": 4.306513786315918, "identity": 0.6342248916625977, "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_AUTH_reg/e55673dfbbb7c0ac5a76f501f23424bcf2298efd8eff183f68bdde6e1b165b68.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "d445d7139880ee45e894bc9a9fddeb2613d19ee1d89954533b90e2698680d0f8", "generation_seconds_batch": 19.482396466075443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3964571952819824, "target_z": 1.1406248476000356, "ce": 6.475345611572266, "genuineness": 4.406930446624756, "identity": 0.37605053186416626, "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_AUTH_reg/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "e70ee384b09abd8720b902cc07c608720785d385d4750554d9151409894757a8", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fcc05df8ac9b5535bb222328ff61db14a3e204e826eacfe3e1ceb205e870d5d", "generation_seconds_batch": 25.790932217030786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 6.403861999511719, "genuineness": 4.369274139404297, "identity": 0.4831433892250061, "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_AUTH_reg/e70ee384b09abd8720b902cc07c608720785d385d4750554d9151409894757a8.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": "e79c8d689631668fe577a847e582e1d65fb2fc93c76ce4bad31037e55ed64702", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cadf8c52f6ad7726a1538055f9df131eaa03913654def1e28a43db46181fc28c", "generation_seconds_batch": 16.5093660390703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 6.149696350097656, "genuineness": 4.356721878051758, "identity": 0.6193156242370605, "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_AUTH_reg/e79c8d689631668fe577a847e582e1d65fb2fc93c76ce4bad31037e55ed64702.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "e8322e014cf16e1ffdcb04cb733a5c4b2e4889f4ae19655f8d8a9549461db908", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "703cabdfadd23dba28729a2f21e9c21fbc4d8f1983508000190c210a39d38563", "generation_seconds_batch": 25.16831897792872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2405447959899902, "target_z": 1.9531247830541683, "ce": 6.086154937744141, "genuineness": 4.1998209953308105, "identity": 0.755041241645813, "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_AUTH_reg/e8322e014cf16e1ffdcb04cb733a5c4b2e4889f4ae19655f8d8a9549461db908.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "e84de38cc43acec62816f516afc52952f21cb248e71badb89ba02fcdd3acc3c5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0db090ff03cb2c4e1c47e712e7d29fc75358cd0fc16cf475c2c0977e4d2852ed", "generation_seconds_batch": 24.97611093102023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.062326908111572, "genuineness": 4.149612903594971, "identity": 0.8084757924079895, "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_AUTH_reg/e84de38cc43acec62816f516afc52952f21cb248e71badb89ba02fcdd3acc3c5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "e98328be738b27f0f2d6b4cf5c7402ec8b55e1f66b4b8523c48c79435d659403", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b66cf9e5bcd4911b230042bd007b5ce488aa4dd80a40816c756906b9ab3c4dd", "generation_seconds_batch": 19.151382499025203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.956477165222168, "target_z": 1.6796876048870344, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.7134420275688171, "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_AUTH_reg/e98328be738b27f0f2d6b4cf5c7402ec8b55e1f66b4b8523c48c79435d659403.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "e987987ff848db3d47df8c0e90021c5b29ced28b3d8fc99fea21110d6ab99c68", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3576fd4b4d94fefc3f60c97789e5c1268dc8067409fc5be87c4d7033a7df823e", "generation_seconds_batch": 25.962707411963493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.157639026641846, "genuineness": 4.149612903594971, "identity": 0.2150583565235138, "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_AUTH_reg/e987987ff848db3d47df8c0e90021c5b29ced28b3d8fc99fea21110d6ab99c68.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "e9b85a7578cf71ef6fbdd2eaf5d818e990fef388a6f7e1af493cd7512a16f290", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc51b5fc92d6b3354896ea092d19721194afe4246672e6bd9122b5320ba1ef64", "generation_seconds_batch": 18.711568597937003, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.284721851348877, "genuineness": 4.193545341491699, "identity": 0.6061632037162781, "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_AUTH_reg/e9b85a7578cf71ef6fbdd2eaf5d818e990fef388a6f7e1af493cd7512a16f290.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "e9bbeffdbc630d83a1ae3cd53b2032000b14c6f837931e3aaad09b05023ecc64", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c50745ff7f1dd7270095a429c6630b8903fbf957f59bde32acf36ffe5b9bced", "generation_seconds_batch": 20.626020450959913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.252951145172119, "genuineness": 4.13078498840332, "identity": 0.5599830150604248, "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_AUTH_reg/e9bbeffdbc630d83a1ae3cd53b2032000b14c6f837931e3aaad09b05023ecc64.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "3fe97ce0b3dbec5960eb8db40dc56392d94fb3194d2ea831f5d103b8a7a37e57", "generation_seconds_batch": 19.897955678985454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.3641486167907715, "genuineness": 4.519899368286133, "identity": 0.4041142165660858, "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_AUTH_reg/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.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": "ea0f3d6521ccfd95618a1f0c5ec3e4956270ea2af29d3fbc19adab57c05a3e1f", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5afc430833910983bed481a46538aea3537eaf0275dc8b7f82a7b19d6795727", "generation_seconds_batch": 22.183417661930434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5651941299438477, "target_z": 2.2656250053788223, "ce": 6.316492557525635, "genuineness": 3.6475296020507812, "identity": 0.5218760371208191, "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_AUTH_reg/ea0f3d6521ccfd95618a1f0c5ec3e4956270ea2af29d3fbc19adab57c05a3e1f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "ea39344cc6036daadfdd359fb79ac2428aa32aebbd44f6eeced7a9e2f53c711f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13b336b50477d8f97662b96677025981ee2208ea137d9b91b45bfc7db1529e75", "generation_seconds_batch": 26.145695246988907, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9077796936035156, "target_z": 1.6328125026894111, "ce": 6.403861999511719, "genuineness": 4.331617832183838, "identity": 0.6180973052978516, "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_AUTH_reg/ea39344cc6036daadfdd359fb79ac2428aa32aebbd44f6eeced7a9e2f53c711f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "ea48c4b669b83a3a53d7517642fc2b4d4641fcc55fd720887656e334a693d29e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b562f35cecad683fbdfd074f4e569ab11d3211223e6704ef4df547ba2ce8d518", "generation_seconds_batch": 26.264507220941596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.046441555023193, "genuineness": 4.036643981933594, "identity": 0.6419275999069214, "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_AUTH_reg/ea48c4b669b83a3a53d7517642fc2b4d4641fcc55fd720887656e334a693d29e.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": "eaaf6dfa57a3a6b73aee001a04040e51814c9687fe18442e447154f41b240a1a", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fa7b2a7b4d09b644d35d087d5a994225afabb97864a0f1449add048f86ad7c2", "generation_seconds_batch": 22.183417661930434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.273009777069092, "target_z": 1.984374851185917, "ce": 6.459460258483887, "genuineness": 3.886019229888916, "identity": 0.37155604362487793, "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_AUTH_reg/eaaf6dfa57a3a6b73aee001a04040e51814c9687fe18442e447154f41b240a1a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "ead7d6508e2c66ecefee21a6e3dd7efc8a0f1f01ddeda7285f5657a7dd0bea1c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4aab6da1b9d163363e62ac7072899c09ea1b4c413792b71bc756acbff39eac6", "generation_seconds_batch": 20.00900287204422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.252951145172119, "genuineness": 4.055472373962402, "identity": 0.6718101501464844, "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_AUTH_reg/ead7d6508e2c66ecefee21a6e3dd7efc8a0f1f01ddeda7285f5657a7dd0bea1c.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": "eb9bcd4b85a555bd984ee5bd5896b70ccac7dffb727f671add9b7c95e46c7ef6", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fe1c91cbabcf733cbcec4ba443c24badc2d3d5a751d4324b393fe70442ad065", "generation_seconds_batch": 16.485854727914557, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.229123115539551, "genuineness": 4.137060642242432, "identity": 0.6308203339576721, "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_AUTH_reg/eb9bcd4b85a555bd984ee5bd5896b70ccac7dffb727f671add9b7c95e46c7ef6.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "eba7c2a6f84ec26d90acb2d927c383aa2d5da456667472b7b079681fc9ec0850", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "96cfda022d57308058903a448b92e388f5fbd447a5b398645e4d594af20d5bb0", "generation_seconds_batch": 25.296295181964524, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149562358856201, "target_z": 2.8281248547717985, "ce": 6.09409761428833, "genuineness": 4.055472373962402, "identity": 0.6187154054641724, "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_AUTH_reg/eba7c2a6f84ec26d90acb2d927c383aa2d5da456667472b7b079681fc9ec0850.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "ebbbd67103aea793b09b5501d5bb71f49e4467445746a4d1451d9e6adb017b53", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f70851392bb53a08888bc1deb36f23ac2b5e15f621832e44b4bc3f9b0255dd6a", "generation_seconds_batch": 20.087681350996718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3152952194213867, "target_z": 1.0625001362634976, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.710437536239624, "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_AUTH_reg/ebbbd67103aea793b09b5501d5bb71f49e4467445746a4d1451d9e6adb017b53.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c28d7a693fb7f69121fdaad85f11bdc980338a1e739e8aaa19d84ab436ea1d6f", "generation_seconds_batch": 18.812717793043703, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8185012340545654, "target_z": 1.546875044823519, "ce": 6.109982967376709, "genuineness": 4.306513786315918, "identity": 0.819063127040863, "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": 0.03999999999999915, "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_AUTH_reg/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "ecb0cc23e64db2eb43688ebfdb2531687f956eb2da96492aa6211687c268e722", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c610f60c0c513b6afd1c194287a523c11cd542499ba75b2ca5db2627001f0d5d", "generation_seconds_batch": 27.127703204983845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.769804000854492, "target_z": 1.5000001721223128, "ce": 6.0305562019348145, "genuineness": 4.256305694580078, "identity": 0.6229878664016724, "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_AUTH_reg/ecb0cc23e64db2eb43688ebfdb2531687f956eb2da96492aa6211687c268e722.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "edcd6a05fd226fa24fdff874d701846be0633c10c6a293176940c8ea450b5e0d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0bb4fc75a7a481aae908670b90f20392ee28feceeab10142d8337f7c9ef0faae", "generation_seconds_batch": 25.0199925879715, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866372108459473, "target_z": 2.093750089647038, "ce": 6.062326908111572, "genuineness": 4.300237655639648, "identity": 0.48886600136756897, "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_AUTH_reg/edcd6a05fd226fa24fdff874d701846be0633c10c6a293176940c8ea450b5e0d.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "ee609eb96f22a3a49811803b71601e39f33f41c3807023e59b74e984036a5cbd", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6b1cf4abe0f84c2fa6cec2cdc777c60944c8fbb0c9930e499b127fa70091e58", "generation_seconds_batch": 26.20526841294486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.213237762451172, "genuineness": 4.300237655639648, "identity": 0.5825866460800171, "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_AUTH_reg/ee609eb96f22a3a49811803b71601e39f33f41c3807023e59b74e984036a5cbd.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "ee6aa60c0ecce9d6c1568e79528269c4b8d60c800ec846999e70a40617534097", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65f8d5d75e4c673940eb6e332cd9937a0964492f6c8ff89b31faa57633e828c7", "generation_seconds_batch": 28.09765619703103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7112860679626465, "target_z": 2.4062498529788576, "ce": 6.086154937744141, "genuineness": 4.193545341491699, "identity": 0.6079542636871338, "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_AUTH_reg/ee6aa60c0ecce9d6c1568e79528269c4b8d60c800ec846999e70a40617534097.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": "ee90a85cb46fd02bdb5c68ad193d55f4cd71b545c96c0c3ca9a718839c6817f0", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95fb0f9f07bdf948bdff056262aca8ffd122db23fcb1ef609f00332dd234a1b3", "generation_seconds_batch": 16.46590609604027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.2767791748046875, "genuineness": 4.231201171875, "identity": 0.4962250590324402, "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_AUTH_reg/ee90a85cb46fd02bdb5c68ad193d55f4cd71b545c96c0c3ca9a718839c6817f0.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "ef595950f9b162b964805a3b9fcf2a748214d1d9bb362e6192c478f39367e432", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e14f8007999e5ff6ffbd20b2ad7d4a3cf4a4169b7868ba3bf43e4260a35fca8f", "generation_seconds_batch": 20.183363327058032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.443574905395508, "genuineness": 4.093128204345703, "identity": 0.7100616693496704, "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_AUTH_reg/ef595950f9b162b964805a3b9fcf2a748214d1d9bb362e6192c478f39367e432.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "f077ac70d997176fc2d18b1e9526152339726693fe8bde234c3df8690a721ad5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2c9c58a122a5ef1ffdd1da26bb400aaf7b48ef02120fe541000915cf23f0128", "generation_seconds_batch": 27.380912416963838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.532729148864746, "target_z": 2.2343749372470736, "ce": 6.133810997009277, "genuineness": 4.268857479095459, "identity": 0.6038312911987305, "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_AUTH_reg/f077ac70d997176fc2d18b1e9526152339726693fe8bde234c3df8690a721ad5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "f07e8e80af654400df1a3aeca5d589fa87f5cceff1f11de092dcb7e773675372", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3df1146f8ade019d99daae938b9eee14d7e2b4b70bfe1138019e35cef81349a2", "generation_seconds_batch": 25.46088673803024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.971005439758301, "target_z": 2.656249939040014, "ce": 6.062326908111572, "genuineness": 4.036643981933594, "identity": 0.7427456378936768, "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_AUTH_reg/f07e8e80af654400df1a3aeca5d589fa87f5cceff1f11de092dcb7e773675372.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "f0980ab688ff371a7c001ab891a14365c637cb38f1cdc6a23ae44b4320c7a9f5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ddf1d0dca0c99bb5e1e80c11db51a6e8cda0fe1eda81b57ac882120ed48afe8", "generation_seconds_batch": 29.17227515799459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.393049240112305, "target_z": 3.0624999067670804, "ce": 6.260893821716309, "genuineness": 4.111956596374512, "identity": 0.7522331476211548, "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_AUTH_reg/f0980ab688ff371a7c001ab891a14365c637cb38f1cdc6a23ae44b4320c7a9f5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "f0acc9dc37d41cec51c58cad450a355a3dc006308821709e2c6b26d199e7a61c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6abde7c7e38930a14dc58206d06ea47bb66737f21abdde6275352753ac7488c", "generation_seconds_batch": 24.935790853924118, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.229123115539551, "genuineness": 4.13078498840332, "identity": 0.7523360252380371, "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_AUTH_reg/f0acc9dc37d41cec51c58cad450a355a3dc006308821709e2c6b26d199e7a61c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "f13b85c23e653479e0d38894647072ccf9041427f54f771d5bae0fb201bcff8a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2cd483c0619926b2ccbcd6391e30b838f94607caffcfdaba576022b265e9310", "generation_seconds_batch": 27.380912416963838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7275185585021973, "target_z": 2.421874887044732, "ce": 6.24500846862793, "genuineness": 4.362998008728027, "identity": 0.6833825707435608, "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_AUTH_reg/f13b85c23e653479e0d38894647072ccf9041427f54f771d5bae0fb201bcff8a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "f14c63931d17c59d395d2408bbdf426c7455db3ef75e45237483d4eb62da9f69", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "31adb1e36632a3bd7f908f273aa4848744a3c4decd84fb7bf5df0b008889e31e", "generation_seconds_batch": 27.428042290033773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9060754776000977, "target_z": 2.5937498027765167, "ce": 6.340320587158203, "genuineness": 4.400654315948486, "identity": 0.5003863573074341, "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_AUTH_reg/f14c63931d17c59d395d2408bbdf426c7455db3ef75e45237483d4eb62da9f69.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "f1a4e67ce09b75021bc8cb9891635beb8cb69c4a5f5fd95563461a47f7d0405d", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "38f98a2a488826deffab950f6ea00a642a700d1f69e1d54f4a096636c1d889ab", "generation_seconds_batch": 19.2274131529266, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0555758476257324, "target_z": 0.8125000502023413, "ce": 6.348263263702393, "genuineness": 4.469690799713135, "identity": 0.5336334705352783, "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_AUTH_reg/f1a4e67ce09b75021bc8cb9891635beb8cb69c4a5f5fd95563461a47f7d0405d.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "5e9940fe247183b771e51a189db65f16ce1e0343357c4f1dd4223e87fe9dfcf6", "generation_seconds_batch": 18.714035636978224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 6.332377910614014, "genuineness": 4.419482707977295, "identity": 0.6272890567779541, "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_AUTH_reg/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "f286b4915f4f6c280104840eb20a7691fdfe545dd8e4f8803d5b627f554cc755", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9630ce668c8622fba4d621b155d706ac54530a22bf63466c5cad246bbb8a6add", "generation_seconds_batch": 27.640296171070077, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246957302093506, "target_z": 2.921875059167045, "ce": 6.006728172302246, "genuineness": 4.124508857727051, "identity": 0.5943366289138794, "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_AUTH_reg/f286b4915f4f6c280104840eb20a7691fdfe545dd8e4f8803d5b627f554cc755.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": "f28ee0add2471ac1640cf21badb33b60f4476d54e0a9122ff069c2ad72957dbc", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "84edc4d84a986d88ed1c9fcd8b2c8ff14df08219c1a7bdf49b9d8e9ab847f835", "generation_seconds_batch": 18.016439547063783, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.403861999511719, "genuineness": 3.8169827461242676, "identity": 0.42095130681991577, "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": 0.120000000000001, "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_AUTH_reg/f28ee0add2471ac1640cf21badb33b60f4476d54e0a9122ff069c2ad72957dbc.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": "f2bc06d6f8b69010eadb25303c388a5dfab9f6304b578d0517b643c93b3c5b15", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0ffe0f1c832565ba653938b46d4535da8715d2e6228cf4d47000064be369ab6", "generation_seconds_batch": 17.19196807104163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.41180419921875, "genuineness": 3.9425034523010254, "identity": 0.4352394938468933, "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_AUTH_reg/f2bc06d6f8b69010eadb25303c388a5dfab9f6304b578d0517b643c93b3c5b15.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "f34553f37acc47cecfa544e639d4539f548826ccd2c52ce9990eb7145ca062b1", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ad95542c72ab04f6c71c610318e9a371a8e918a82a4e6b8216248a6ce17f7df", "generation_seconds_batch": 25.82176985195838, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3704047203063965, "target_z": 2.0781250555811637, "ce": 5.95112943649292, "genuineness": 4.231201171875, "identity": 0.5976652503013611, "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_AUTH_reg/f34553f37acc47cecfa544e639d4539f548826ccd2c52ce9990eb7145ca062b1.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "f4b42bb6cffedbca72b3c2796b5ce6e58259fa27cc7edad939fe1f71bf8ad924", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6be59b03989de971fe9afe16d66d40943ccf159089160458111b6fb4382c35f", "generation_seconds_batch": 18.845960006932728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.427689552307129, "genuineness": 4.042920112609863, "identity": 0.622075617313385, "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_AUTH_reg/f4b42bb6cffedbca72b3c2796b5ce6e58259fa27cc7edad939fe1f71bf8ad924.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "f519b44fb02264990e6ff6adf28f30665f083e494d787af2462c30c139738454", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4acd4db8a4d74d34a8955523d9159da5f2b9314a411750f861b9b1ce204e864c", "generation_seconds_batch": 20.897666932898574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.95477294921875, "target_z": 2.64062490497414, "ce": 6.109982967376709, "genuineness": 4.193545341491699, "identity": 0.7044925689697266, "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_AUTH_reg/f519b44fb02264990e6ff6adf28f30665f083e494d787af2462c30c139738454.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "f5a5f4f294b1941ab645d95f808e9322c5e3e5ee23c76d05c1184df09f51b18e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdce9998876a3b1c53e4abb16f46acdcf0b7a937226180530c1eae1dfe0de435", "generation_seconds_batch": 24.270862029050477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3802249431610107, "target_z": 1.1250000430305782, "ce": 6.1020402908325195, "genuineness": 4.469690799713135, "identity": 0.6694102883338928, "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_AUTH_reg/f5a5f4f294b1941ab645d95f808e9322c5e3e5ee23c76d05c1184df09f51b18e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "f5c443a462bb20ead755ae0cb145f4802c8a6bd2e18b99b377d5a2185e01e2aa", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "48764c651036c98765f3b6903eceea1d8c651b754d98d29eb561f054baced369", "generation_seconds_batch": 20.49468479701318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2665977478027344, "target_z": 1.0156250340658743, "ce": 6.403861999511719, "genuineness": 4.231201171875, "identity": 0.5186249613761902, "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_AUTH_reg/f5c443a462bb20ead755ae0cb145f4802c8a6bd2e18b99b377d5a2185e01e2aa.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": "f5fc60b2617c70eaca35544690fed2b5eac4045394588c1148651b3b56bd295e", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5fc3049f87af487012a7abdd294012574e327483e4832066eeaf63b36669e370", "generation_seconds_batch": 16.58228388905991, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.165794849395752, "target_z": 2.843749888837673, "ce": 6.1894097328186035, "genuineness": 3.8420867919921875, "identity": 0.6159220933914185, "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_AUTH_reg/f5fc60b2617c70eaca35544690fed2b5eac4045394588c1148651b3b56bd295e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "f61f81705e4f3b2100788981985808ee2821a5e1cb4e0ba08177cedecf3ad421", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eed3d930e83b6a86c2c22f9ab2c59a23fa940d278f8ca5e1ddc94ac93394fb8a", "generation_seconds_batch": 25.666995398001745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.102569103240967, "target_z": 1.82031245248707, "ce": 6.268836498260498, "genuineness": 4.080576419830322, "identity": 0.3471020460128784, "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_AUTH_reg/f61f81705e4f3b2100788981985808ee2821a5e1cb4e0ba08177cedecf3ad421.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": "f717cd7dbae7fac8cdf47cd9581d157c62d6582fd9075d290e10ac0aec23a105", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "792c2e2f7a9a857c82b7ccd8f15fe3ebea9ae27d2dfae5e23f45df2b9c424df8", "generation_seconds_batch": 25.69512859499082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 5.998785495758057, "genuineness": 4.111956596374512, "identity": 0.7313525080680847, "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_AUTH_reg/f717cd7dbae7fac8cdf47cd9581d157c62d6582fd9075d290e10ac0aec23a105.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": "f79db0f155737066ae28e11a58f11bb7f5a6642163fe19b9226181e1297c94f0", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6aad98b6c745245f75a32259e688072c14c26a8c71db5bb3270e2d2f36a79dda", "generation_seconds_batch": 16.100515394937247, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6138916015625, "target_z": 2.3125001075764455, "ce": 6.23706579208374, "genuineness": 4.268857479095459, "identity": 0.5805840492248535, "hypothesis": "Deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann lagte 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_AUTH_reg/f79db0f155737066ae28e11a58f11bb7f5a6642163fe19b9226181e1297c94f0.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "6afa6bec9ce3a04259acc57fcc1efded3f0385a5189e32277bb46544d1083403", "generation_seconds_batch": 22.326247585006058, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.623711585998535, "target_z": 1.359374865529443, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.7065322399139404, "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_AUTH_reg/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "f7b5ea1bce3e5311d02ee2f51a686acac2f1848f19d9c635e913f1bfd5e7b15c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06d28840c433f51b71b7d9312adddd179bdac905fd347271d20b61d0b9107b7c", "generation_seconds_batch": 28.828807416022755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.348263263702393, "genuineness": 4.287685871124268, "identity": 0.4011451005935669, "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_AUTH_reg/f7b5ea1bce3e5311d02ee2f51a686acac2f1848f19d9c635e913f1bfd5e7b15c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "f7dc940142e7245485d17cdc771a43f78275479c7bb34d938879b919a90a9784", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f6de69f89c9018e71e8c5ae361f70f07a67411b9f71eab6295816e18fae4e05", "generation_seconds_batch": 25.710229006945156, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4840316772460938, "target_z": 2.1874998350494503, "ce": 6.205295085906982, "genuineness": 3.8671910762786865, "identity": 0.40991851687431335, "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_AUTH_reg/f7dc940142e7245485d17cdc771a43f78275479c7bb34d938879b919a90a9784.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "f7fc3371540de8250223fbb23f57ddd328c6188e760cdfab0ad76f82592477f3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "732c4bd65a9a2020b2bcddbc897c9cbf083821a4d71d493e3efd183f97a43428", "generation_seconds_batch": 25.609792027040385, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 6.125868320465088, "genuineness": 4.13078498840332, "identity": 0.6167538166046143, "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_AUTH_reg/f7fc3371540de8250223fbb23f57ddd328c6188e760cdfab0ad76f82592477f3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "f8324c71153a32b68495955b9fa87f147d618016c32d55d15c6e8544206e7517", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a455ce1fc403fa465a738ea4d9a8c68d817b7ea40f0e00a61bbbb4ad245aef10", "generation_seconds_batch": 26.0075959690148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6480603218078613, "target_z": 1.3828124166282547, "ce": 6.268836498260498, "genuineness": 4.532451152801514, "identity": 0.7045551538467407, "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_AUTH_reg/f8324c71153a32b68495955b9fa87f147d618016c32d55d15c6e8544206e7517.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": "f83dd948f1572c645ebd42144ef12270a49d8eee8ca308b84cd9e40db7c63ec9", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f83b0f52a04f8acb56b572e6d95c43f0b0eeef71b90afd4881a32a2ba6872a23", "generation_seconds_batch": 22.128475266043097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.474211692810059, "target_z": 3.1406250770964528, "ce": 6.284721851348877, "genuineness": 4.13078498840332, "identity": 0.49117475748062134, "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_AUTH_reg/f83dd948f1572c645ebd42144ef12270a49d8eee8ca308b84cd9e40db7c63ec9.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "f84775273a603aba4e7dab5ebc8a943b85fdaa1f69746db5058101a61640fef3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a2a23b618a3a05b0fe3bc9eefd43d0a018a4148a5089c18d9b13ffed044b6caa", "generation_seconds_batch": 20.834412113996223, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.8016366362571716, "hypothesis": "J'ai trouvé la petite boîte, tapoffer la bijudu, elle était près de la vieille fenêtre, et morchée du sonore.", "wer": 0.5263157894736842, "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_AUTH_reg/f84775273a603aba4e7dab5ebc8a943b85fdaa1f69746db5058101a61640fef3.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_AUTH_reg", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": "dbb0d3f6db905752167d740719f7d44a7a32f021857561945974687317c9689d", "generation_seconds_batch": 19.793420445057563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7373390197753906, "target_z": 1.468750103990564, "ce": 6.197352409362793, "genuineness": 4.406930446624756, "identity": 0.5835305452346802, "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_AUTH_reg/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.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": "f8dadc844e2715c7683386f07e611fe379a75c7626fb670d1c282262f71c4a4e", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73725619a1ffce98268020c0289e7231f91443e07f6a1f15d6dcffda9554e7e1", "generation_seconds_batch": 22.155357246054336, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13332986831665, "target_z": 2.812499820705924, "ce": 6.308549880981445, "genuineness": 4.231201171875, "identity": 0.18246051669120789, "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_AUTH_reg/f8dadc844e2715c7683386f07e611fe379a75c7626fb670d1c282262f71c4a4e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "f97fcde37cca3f1cfb4bf8743fc02b7aa34aa1ab3da82b69db5e38266865314f", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b46fda0b17e6268827630cc9c0a19f1ad23dbe2fb330c55ca65b20f1cdf72e8e", "generation_seconds_batch": 24.797879952937365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.003470420837402, "target_z": 2.687500007171763, "ce": 6.332377910614014, "genuineness": 4.231201171875, "identity": 0.5008839964866638, "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_AUTH_reg/f97fcde37cca3f1cfb4bf8743fc02b7aa34aa1ab3da82b69db5e38266865314f.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "f99ae238804398ba2baf766c98b979b786c8bb0c0767a81f34e8b2a486a9023a", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "654c318569df30c8441a028b97631415557a125b58e67b691d2771b93166cb3f", "generation_seconds_batch": 27.38550260500051, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.06198787689209, "target_z": 1.7812498673223838, "ce": 6.348263263702393, "genuineness": 4.6203155517578125, "identity": 0.41227197647094727, "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_AUTH_reg/f99ae238804398ba2baf766c98b979b786c8bb0c0767a81f34e8b2a486a9023a.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-0", "key": "f9c7ca422e52c57284deab1004efff52b303badbb8dc153a07f290345f5cefc5", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9ac4be4db2bd03e3d07a7bb2f1e8dafed7dd1fc8a2a34a82e3df58139a4278ee", "generation_seconds_batch": 19.45925772492774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.808681011199951, "target_z": 2.500000057374104, "ce": 6.0146708488464355, "genuineness": 3.9927117824554443, "identity": 0.6883686780929565, "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_AUTH_reg/f9c7ca422e52c57284deab1004efff52b303badbb8dc153a07f290345f5cefc5.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-fr-1", "key": "fa514ee7925e3f0cd4e34e83d4329a9df9b2dfea7e7fc64ff95e204b5b911ffe", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8773c8155667974364a5d9eec3ccafc9b2aed22d3956a860c845b2497089190", "generation_seconds_batch": 20.087681350996718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 6.0305562019348145, "genuineness": 4.406930446624756, "identity": 0.7725996971130371, "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_AUTH_reg/fa514ee7925e3f0cd4e34e83d4329a9df9b2dfea7e7fc64ff95e204b5b911ffe.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": "fa6f286326a164166b20565b5b8db9c10af3507f0438415a0ca2623d865c96e7", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdbf732e006572e15017fa7f090c31e9ab61b57911d996f38860c22632182fe5", "generation_seconds_batch": 26.45291008404456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.100865364074707, "target_z": 2.7812502115670097, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.6562063097953796, "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_AUTH_reg/fa6f286326a164166b20565b5b8db9c10af3507f0438415a0ca2623d865c96e7.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "fab19d879c758d2e7b240eb1cc732595a7e912e4c1293d23905b9bc3b1f7b002", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "412733c8d489bfc140d7100320baacc9c5039f59656fd4f357b26a0ffe9c464b", "generation_seconds_batch": 25.422244618996046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.743751049041748, "target_z": 2.4374999211106068, "ce": 6.141753673553467, "genuineness": 4.099404335021973, "identity": 0.55492103099823, "hypothesis": "Die Tür ist noch ist offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "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_AUTH_reg/fab19d879c758d2e7b240eb1cc732595a7e912e4c1293d23905b9bc3b1f7b002.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "fabb39dcdc15a5018854e3ae5c29583ef0055c43bbc384ad962b0aaad535dc6e", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a2f71f2e86777d28b9eefbad0c67ccfd248283a6d6158ba0c384d0b33fd5111", "generation_seconds_batch": 18.99400839698501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.940244674682617, "target_z": 1.66406257082116, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.5829198360443115, "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_AUTH_reg/fabb39dcdc15a5018854e3ae5c29583ef0055c43bbc384ad962b0aaad535dc6e.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-1", "key": "fb3cd5429d15cf8317bf50dc07eee43b65be5b36e0eed6a3ac938a0d1d2e6aa3", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ef5c77e9c9ace406f37f902043a9eafd56ae26000d5edb37bd77c571b20cdd34", "generation_seconds_batch": 24.335674518020824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 5.597680568695068, "genuineness": 4.250029563903809, "identity": 0.5261399149894714, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.05555555555555555, "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_AUTH_reg/fb3cd5429d15cf8317bf50dc07eee43b65be5b36e0eed6a3ac938a0d1d2e6aa3.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "fb7d2e05226b23292ee98eade619749630fad4285e0547f4846db3f723d61bdc", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "320210e81ca40b74c9e3acf51f7d8aaf4b4cc2f8eb4880df29b8002aa2ad1cad", "generation_seconds_batch": 26.280841688043438, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.938540458679199, "target_z": 2.6249998709082654, "ce": 5.95112943649292, "genuineness": 3.898571252822876, "identity": 0.6467899680137634, "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_AUTH_reg/fb7d2e05226b23292ee98eade619749630fad4285e0547f4846db3f723d61bdc.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-1", "key": "fbd30ff16d877f10087bc7c2d22f7d539e4e68519464ff94d3ba9d235a125536", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f987e5df1a4632e0eafefb164c1d77b1b8adaeea2f40ee6bd79935fcb97402aa", "generation_seconds_batch": 26.72364959702827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5164966583251953, "target_z": 2.218749903181199, "ce": 6.316492557525635, "genuineness": 4.093128204345703, "identity": 0.6552233099937439, "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_AUTH_reg/fbd30ff16d877f10087bc7c2d22f7d539e4e68519464ff94d3ba9d235a125536.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-1", "key": "fc442aa64d078727203955d4cc84e8874013888fbae392130cf6b8aebac5a108", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] I found the little box.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22930aa5c2537fc444e39248c7e1ab587d0230686074c1ef1621c01056a87138", "generation_seconds_batch": 19.221954049076885, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2909464836120605, "target_z": 1.039062585164686, "ce": 6.332377910614014, "genuineness": 4.193545341491699, "identity": 0.5428463220596313, "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_AUTH_reg/fc442aa64d078727203955d4cc84e8874013888fbae392130cf6b8aebac5a108.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "fd35829f77413661d220406842087a02b30e2a74f410e6a66b12cb901c99d63b", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cd62e39fde7375a75e94eb3b3d76fb19d50c841d8e23ce70ece83e818c2478f7", "generation_seconds_batch": 27.43370363500435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6625890731811523, "target_z": 2.359375209774069, "ce": 6.205295085906982, "genuineness": 4.017816066741943, "identity": 0.16219231486320496, "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_AUTH_reg/fd35829f77413661d220406842087a02b30e2a74f410e6a66b12cb901c99d63b.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-0", "key": "fd5d1f9997338ff8eb169dc4b71838756dab49c0c0fd9a7bebaa131fdae2585c", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a21445235d86a0551ca071843d5ba03ea5fcb00b39fb92a281659cd654597e88", "generation_seconds_batch": 25.05696232803166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3541722297668457, "target_z": 2.062500021515289, "ce": 5.919358730316162, "genuineness": 3.9299514293670654, "identity": 0.6942875385284424, "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_AUTH_reg/fd5d1f9997338ff8eb169dc4b71838756dab49c0c0fd9a7bebaa131fdae2585c.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-es-1", "key": "fda648f5c31330c8aae2994dd841174c9f5be7bd9a0be29dd043d7b5589a8665", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5487217e1e9ea28d9ed6834a3bb09877d409e1880510066c51981f8610e557f9", "generation_seconds_batch": 17.86445732298307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3054747581481934, "target_z": 2.015624919317666, "ce": 5.927301406860352, "genuineness": 4.250029563903809, "identity": 0.7333927154541016, "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_AUTH_reg/fda648f5c31330c8aae2994dd841174c9f5be7bd9a0be29dd043d7b5589a8665.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-en-0", "key": "fdd780d488543ed7a155ee6f7030c7fd9ec952d2b8be745f2e6f61a4d6556432", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] The door is still open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d9b894626cacb960597d65dcd578eb7f3d19fab183fc5084b176bbacebbdc21", "generation_seconds_batch": 24.811311046010815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.72110652923584, "target_z": 1.4531250699246896, "ce": 6.157639026641846, "genuineness": 4.256305694580078, "identity": 0.3637193739414215, "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_AUTH_reg/fdd780d488543ed7a155ee6f7030c7fd9ec952d2b8be745f2e6f61a4d6556432.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-nl-0", "key": "ff58d2ff3abcd0439d0f13f0f240bd687c0cc0d01b74550b3dbb632cffdf73ce", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6fb4b24bba580aa4223656fc2d4f5e2fbe7498b015c1fbb80d1c759951951577", "generation_seconds_batch": 24.66350451495964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.425514221191406, "target_z": 3.0937499748988295, "ce": 6.062326908111572, "genuineness": 3.9927117824554443, "identity": 0.5087735652923584, "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_AUTH_reg/ff58d2ff3abcd0439d0f13f0f240bd687c0cc0d01b74550b3dbb632cffdf73ce.mp3" }, { "id": "vn_S_AUTH_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_AUTH_reg-de-0", "key": "ff8abb0f28c1e7554f6a8de3978bf3ccdd3379267fbe62d8a7dfa3daadc7ae50", "bucket": "vn_S_AUTH_reg", "phase": "grid", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery 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": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2ae9888828a5114ad174f6337fbb19df09833de1a92d8342a3a1f5cb4c459c8", "generation_seconds_batch": 25.326602155924775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.356205940246582, "genuineness": 4.143336772918701, "identity": 0.43588191270828247, "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_AUTH_reg/ff8abb0f28c1e7554f6a8de3978bf3ccdd3379267fbe62d8a7dfa3daadc7ae50.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": "ffcc046803a24bd05411e369ba00f6b8bbf4587b59394f74cd3d335e29b7c3b2", "bucket": "vn_S_AUTH_reg", "phase": "heldout", "candidate": "manual-553924db97fc", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Establish it immediately. Let the local direction carry the intended delivery beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Sustain the delivery. Let the local direction carry the intended delivery beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Calm, unmistakable authority; grounded resonance, firm attacks and decisive sentence endings. Preserve the intended style through the end. Let the local direction carry the intended delivery beside each sentence.) [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": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "892e76ad5a9f52b63a978062280e7145f2f4ae92df4b8056421d13aaf8c73e35", "generation_seconds_batch": 25.952241010032594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.019702911376953, "target_z": 2.7031250412376373, "ce": 6.157639026641846, "genuineness": 4.099404335021973, "identity": 0.6510635614395142, "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": 0.120000000000001, "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_AUTH_reg/ffcc046803a24bd05411e369ba00f6b8bbf4587b59394f74cd3d335e29b7c3b2.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "3051a8941685c341fe5c638e1b586f7f09d5880898c029272b9cca49fc72b17b", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.7292227745056152, "target_z": 1.4609375869576269, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.46256837248802185, "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_AUTH_reg/de/fe768d0ffb574f263c62d3bd.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a50423e30bc966293dca8a64ff2abee994cbfa4a5abc2495c3064d6163d4f700", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8996634483337402, "target_z": 1.6249999856564739, "ce": 6.324435234069824, "genuineness": 4.545003414154053, "identity": 0.23980671167373657, "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_AUTH_reg/de/fd7ac0f19c607153833d21eb.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "9a3d6bff1683653aae9e6507c313a36700a5b487317c9d35321db556b200c7df", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.133810997009277, "genuineness": 4.118232727050781, "identity": 0.4368104338645935, "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_AUTH_reg/de/825a2c304b8f09fd81766c25.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "e596f48bb9dc221a593d779c48e970b917e5e953d54b4ebf826b428c942332e0", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.24500846862793, "genuineness": 4.344170093536377, "identity": 0.490317165851593, "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_AUTH_reg/de/b53d41c930bdaa658dc86014.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "23ba1f1448c172d8e1b2f7986ca1dab4b0d505b1c42a7ea9d517b7dfa4e10cf8", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.1756153106689453, "target_z": 1.8906251057835048, "ce": 6.292664527893066, "genuineness": 4.469690799713135, "identity": 0.29712000489234924, "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_AUTH_reg/de/2259612646bc576297e14f06.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "eac78c0f870efa1a16dce76a98e05886d9dcfe9948924c178905df62a6142ab6", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.4353346824645996, "target_z": 2.140625191844661, "ce": 6.141753673553467, "genuineness": 4.2249250411987305, "identity": 0.43825826048851013, "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_AUTH_reg/de/d5b4581c9bdceb15a17831f3.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "431d47569e29a4c31955a4b2fb84482049473b8009d5f1695dcacc6df56089a3", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8671984672546387, "target_z": 1.5937499175247252, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.49549904465675354, "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_AUTH_reg/de/4798a9639c78cb2e021d0a03.mp3" }, { "key": "original-vn_S_AUTH_reg-de-0-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "9cbb30ea7cf0cf76c573da0d78ba167d872da5f08918371ca7c1da484ef8da9f", "task_id": "vn_S_AUTH_reg-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.5242562294006348, "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_AUTH_reg/de/2f28cc0e3661d1c782eac179.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4f1d0df310414dc0f23be2d8e5b33540241d245093fbbc019f7ebf4e426e508d", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.005174398422241, "target_z": 1.7265624775882404, "ce": 6.3641486167907715, "genuineness": 4.1998209953308105, "identity": 0.609660267829895, "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_AUTH_reg/de/9079d4652e159c00338bf2fc.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "0b2ba6e82cf345114587b3f16e732da23abe059b08e8362117608b6155d4e23b", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.5002641677856445, "target_z": 2.2031248691153245, "ce": 6.395919322967529, "genuineness": 4.1998209953308105, "identity": 0.625805139541626, "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_AUTH_reg/de/5305945df560e23b44c4880c.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "665693ab746e35ca5a5fa234238347e97dc57df692bf3c8ac4ffe2987a1f732b", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.5814266204833984, "target_z": 2.281250039444697, "ce": 6.395919322967529, "genuineness": 4.080576419830322, "identity": 0.6050732731819153, "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_AUTH_reg/de/65d098ceadf28726ac42b7ce.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "7485120389d38138db09e97a7c82b4f9b3049577a1f428142fa0dbf650b221b2", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.356205940246582, "genuineness": 4.218649387359619, "identity": 0.5565690994262695, "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_AUTH_reg/de/3fb2aa1204bcd493257cac27.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "042f1a0e58a08e1c5fbc540a1503256efa23d3231d37c963505b6ce296054eb3", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.5912468433380127, "target_z": 1.3281250268941114, "ce": 6.372091293334961, "genuineness": 4.582659721374512, "identity": 0.400684654712677, "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_AUTH_reg/de/9f824b90cf413cec09f54849.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "cf3fa6e7ebbdce26317345d6f1cc3ddb949da5c3095287ae095f5a0f282972ab", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.980825901031494, "target_z": 1.703125155985846, "ce": 6.41180419921875, "genuineness": 4.243753433227539, "identity": 0.48287010192871094, "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_AUTH_reg/de/b113b0df8e97aa78243d82bd.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "16932b3d468c7b7be42dd09fe595b049a57b666ae732c1e7471137f75639867b", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.81038498878479, "target_z": 1.5390625277905818, "ce": 6.38797664642334, "genuineness": 4.356721878051758, "identity": 0.5238480567932129, "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_AUTH_reg/de/dcbe119f56a41ab7cecc9024.mp3" }, { "key": "original-vn_S_AUTH_reg-de-1-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a5292a363d2c89ccdc123a1a85ca8c155bce538048129852afd1392e6c80131d", "task_id": "vn_S_AUTH_reg-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.0295231342315674, "target_z": 1.750000028687052, "ce": 6.427689552307129, "genuineness": 4.444586753845215, "identity": 0.32025307416915894, "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_AUTH_reg/de/1202317b08c23ec908af8438.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "d6108b626df55d3ccf19c3470c1ccf239fde65e5152401ec5050e65d7e09ed71", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.428922176361084, "target_z": 1.1718749157317843, "ce": 6.205295085906982, "genuineness": 4.243753433227539, "identity": 0.1743023693561554, "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_AUTH_reg/en/1a67eb05581f63a16e44b0a5.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "eee71b9547d9ec1d2fc3bbb676550e65da0e5641fcc4b499f25e56cfd9d50cc4", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.1431503295898438, "target_z": 1.859375037651756, "ce": 6.23706579208374, "genuineness": 3.9173994064331055, "identity": 0.29459887742996216, "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_AUTH_reg/en/888edc3e63e5aea72e131ef5.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a176aa4807399c9963ea6656b6a5d1f128bbfe6a0887658df5d80687b5d072fd", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.126917839050293, "target_z": 1.8437500035858816, "ce": 6.1894097328186035, "genuineness": 3.7291181087493896, "identity": 0.4011373519897461, "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_AUTH_reg/en/e57621a252310565586d3161.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "06632420fc431d38c5b7ca0f6b4a4d8d93a798aa8163243a879b517500942466", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.3802249431610107, "target_z": 1.1250000430305782, "ce": 6.252951145172119, "genuineness": 4.143336772918701, "identity": 0.432208389043808, "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_AUTH_reg/en/dcbccdccec860b0303e82ec9.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "705a90101395d97d2f716c220ab8a5800c982bb33ceea07e1e0c14bcf8449328", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.232429027557373, "target_z": 1.9453127250140652, "ce": 6.229123115539551, "genuineness": 4.162164688110352, "identity": 0.3000306487083435, "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_AUTH_reg/en/cbe61fa40d27b569c352824e.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "6815f08f34c1b332880eb95ab8eed62874bc3f2c4f856c4b87085594bb856460", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.1976099014282227, "target_z": 0.9492188687823253, "ce": 6.252951145172119, "genuineness": 4.356721878051758, "identity": 0.3022615909576416, "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_AUTH_reg/en/59f9c52a0809bf1fac308f75.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "b311173fd6b33f877c95038bdb8aadb3aab66513a1b45ce884fa97fd087b54a6", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.4515671730041504, "target_z": 2.1562502259105356, "ce": 6.24500846862793, "genuineness": 3.8169827461242676, "identity": 0.20736369490623474, "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_AUTH_reg/en/462c4e812cb79f1a74f69034.mp3" }, { "key": "original-vn_S_AUTH_reg-en-0-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "0d0635a6455c2afc046caba8afba8ff9dfb72722f8df151c30870c2e5681c896", "task_id": "vn_S_AUTH_reg-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.321707248687744, "target_z": 2.0312499533835404, "ce": 6.173524379730225, "genuineness": 3.9425034523010254, "identity": 0.2794348895549774, "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_AUTH_reg/en/4bbe8eb7a16ed4aafc95c46f.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a1a682e9e35aecc6fc39ae81cba6806e1f4eeddc0447e79646d06b6ca2e49563", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.8445539474487305, "target_z": 0.6093750663388081, "ce": 6.3641486167907715, "genuineness": 4.507347106933594, "identity": 0.38668161630630493, "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/audio/vn_S_AUTH_reg/en/b53cb769dc19488d0cea990b.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "2dcc32c9a4419349fa90590c4658bc0657f082bbe4674092d9447e62903e9219", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.7877402305603027, "target_z": 0.5546874471082476, "ce": 6.292664527893066, "genuineness": 4.344170093536377, "identity": 0.3179980516433716, "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_AUTH_reg/en/a79714dc39ad059ae2627970.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "80a0194877d7478a4d8e1a468bd4e5169899b556d633ca16bf7efcbb15f29d2a", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.2747139930725098, "target_z": 1.0234375510988116, "ce": 6.427689552307129, "genuineness": 4.231201171875, "identity": 0.34721478819847107, "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/audio/vn_S_AUTH_reg/en/d9952c48e7ff006dcd1bdad9.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "fd5c13fa94f15c22c939ec94dad5af9fbb9f324a920f99a40d4e66ce0adecace", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.9257161617279053, "target_z": 0.6875000071717631, "ce": 6.252951145172119, "genuineness": 4.243753433227539, "identity": 0.4132642447948456, "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_AUTH_reg/en/dc533c965d73ac2dd59d7b8e.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "99b1abef17e909a7616a6fc3036eb3a8e77a7a8b7892c545f026c70f3e9058b5", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.124563694000244, "target_z": 0.8789062154858904, "ce": 6.395919322967529, "genuineness": 4.444586753845215, "identity": 0.300930917263031, "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_AUTH_reg/en/ea98ef18bca6fbee5e64b8cc.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "b6a94844dc260a4db67ecd9ef41f32b56c9f9210e4d8ee5177cba7a61eb105e8", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.8526701927185059, "target_z": 0.6171875833717453, "ce": 6.260893821716309, "genuineness": 4.419482707977295, "identity": 0.3632207214832306, "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_AUTH_reg/en/27f09ab83b1c266e18934cb7.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "704893221a3f22c71d91a5bbbbdb90bed2431a663c2cfd00b1513db713154963", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.6152703762054443, "target_z": 0.3886718044029576, "ce": 6.41180419921875, "genuineness": 4.394378662109375, "identity": 0.17565351724624634, "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_AUTH_reg/en/37d05c3156c369d6bbc7d1dc.mp3" }, { "key": "original-vn_S_AUTH_reg-en-1-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "f959e1fa2fc3b605f4717cb9bc7a0c3dde73be96ba1bc4ba4ad2481d88d75601", "task_id": "vn_S_AUTH_reg-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.165144920349121, "target_z": 0.9179688006505764, "ce": 6.324435234069824, "genuineness": 4.2249250411987305, "identity": 0.3616064488887787, "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_AUTH_reg/en/093807e1df20ee7515bd06d5.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4c7e064b40aa818b09162ec5265b08cf295ce4cb32bad57a93a19e63927b6082", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.6074790954589844, "target_z": 1.3437498314635687, "ce": 5.692992687225342, "genuineness": 4.256305694580078, "identity": 0.7908214330673218, "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/audio/vn_S_AUTH_reg/es/9a07220e3acef0d77bf69cdd.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "36155e5bae436812dfd77bfae746302922a7f0bd3a12dc182d8b0d904065734f", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.1999640464782715, "target_z": 1.9140626568823165, "ce": 6.23706579208374, "genuineness": 4.168440818786621, "identity": 0.4982132315635681, "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_AUTH_reg/es/dcef56604eceec9de31aea40.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4bc66efd519bd6ce9ee47adb293337d4a996dbdfc205867da06b4ed3379916ff", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.191847801208496, "target_z": 1.9062501398493792, "ce": 5.9431867599487305, "genuineness": 4.243753433227539, "identity": 0.7392469644546509, "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_AUTH_reg/es/90cdf08c58c9b80a36c959f7.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "f91f29e667c1ce22ed481153b82f6274d4a5b6d3a7687e4cc52cc7437b49436c", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.157639026641846, "genuineness": 4.2751336097717285, "identity": 0.5782305598258972, "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_AUTH_reg/es/0a70b856f977c13efddd0652.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "682a2f202d24fd7199075f717426e83b18631fe562149b0c4b6a57de54827129", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.9727096557617188, "target_z": 1.6953126389529087, "ce": 5.959072113037109, "genuineness": 3.9048471450805664, "identity": 0.5926581621170044, "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_AUTH_reg/es/3b0e5f84fb03352b4dc91fb6.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4fa9b256e31a51e950c43f53848372e947090b65a20f589ac8544e45d1509b98", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.704874038696289, "target_z": 1.4375000358588153, "ce": 5.863759994506836, "genuineness": 4.061748504638672, "identity": 0.4642760753631592, "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_AUTH_reg/es/950c326f3e7dda274fb8da28.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4dfa3fbacfc41a74e3c32ee96579228a3ced0fae828cfb998925a4ca09a6abdd", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 5.557967185974121, "genuineness": 3.9173994064331055, "identity": 0.6847662925720215, "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_AUTH_reg/es/8f2aadb3bc81c60fdac7a9b0.mp3" }, { "key": "original-vn_S_AUTH_reg-es-0-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "8e578933ab8e2b80184d85b1a9bd135ac50a2ad2e2bc169c561e483779cd1448", "task_id": "vn_S_AUTH_reg-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.5182008743286133, "target_z": 1.2578126030940937, "ce": 6.022613525390625, "genuineness": 4.18099308013916, "identity": 0.5260601043701172, "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/audio/vn_S_AUTH_reg/es/55620cadb63fbf1f6a76a47b.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "fedc74762067eee9b99e3d5d26446fbe9df810f431d07cefe543c1ef8a63a917", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.2016677856445312, "target_z": 0.9531248978023767, "ce": 6.062326908111572, "genuineness": 4.7834930419921875, "identity": 0.6137035489082336, "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_AUTH_reg/es/77326fd81a2cff2711ed7dda.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "8ef61c4d6734ee2cdfbd6c23a1ab468af532cb6badf9094c3ae2dbfd0429b9c0", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.915895938873291, "target_z": 1.6406250197223484, "ce": 6.125868320465088, "genuineness": 4.281409740447998, "identity": 0.6304880380630493, "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_AUTH_reg/es/3489af0be6d82750ff8fc6ad.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "65a5fda349f977f29527cec1297fc33c9e3e93f5d4d7fd76834585dcce216110", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.2138423919677734, "target_z": 0.9648439028481998, "ce": 6.324435234069824, "genuineness": 4.7332844734191895, "identity": 0.6472311019897461, "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_AUTH_reg/es/d7e69960af885fe4f97c87f2.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "8680743048db11302aec2f9f37d250b58987d493dd44c4652f7d214e0d5455ba", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.997058153152466, "target_z": 1.7187499605553034, "ce": 6.141753673553467, "genuineness": 4.425758361816406, "identity": 0.5688123106956482, "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_AUTH_reg/es/b88301c8d0d17d5ec33b8eb3.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "54dfd8427713cd19228a59ca6f3934a846a7a82538bb6f6f004f7e5875d9d022", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.347759962081909, "target_z": 1.0937499748988293, "ce": 6.070269584655762, "genuineness": 4.532451152801514, "identity": 0.650895893573761, "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_AUTH_reg/es/d9befda9e0c14208429582d9.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "e1b31fad0827cad3df6233f11bf0bb7bc3821b7aff11bc70cd6990768c04f51c", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.3802249431610107, "target_z": 1.1250000430305782, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.6225863695144653, "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_AUTH_reg/es/66edc2e144d947fa1a343fe0.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "6624db74153078c966d70d5d71b626188da58bc549203aaf0b46a93795fdce2c", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.128621816635132, "target_z": 0.882812474002359, "ce": 6.252951145172119, "genuineness": 4.7458367347717285, "identity": 0.7048338651657104, "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_AUTH_reg/es/d2f7629e2b05471541c79034.mp3" }, { "key": "original-vn_S_AUTH_reg-es-1-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "1473c82e119651075605de231649f8bd164d62c1b525c588e78236a172230572", "task_id": "vn_S_AUTH_reg-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.745455265045166, "target_z": 1.4765626210235012, "ce": 6.1020402908325195, "genuineness": 4.099404335021973, "identity": 0.5693190097808838, "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_AUTH_reg/es/7c46921a5ac66c578b9ad843.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "294e01bbcd4624b5d855e652ebbeff5191982b2eba84f5f5da0446532b1e7c81", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.347759962081909, "target_z": 1.0937499748988293, "ce": 6.109982967376709, "genuineness": 4.243753433227539, "identity": 0.8170621991157532, "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_AUTH_reg/fr/cadf35a85cd16319895e4b76.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "4ad481684222adc08027fb8b188a0ce84330c48fa5d509dce08d2ffe9ba7c253", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.5668981075286865, "target_z": 1.3046874757952998, "ce": 6.022613525390625, "genuineness": 4.331617832183838, "identity": 0.5239301919937134, "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_AUTH_reg/fr/ec97b1e17ec5cfda65ebf46c.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "27646c38d4036724d4a2b0c31a8493fc66e7ca179630de6c97046b71cf548598", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.0109364986419678, "target_z": 0.7695312065211866, "ce": 6.316492557525635, "genuineness": 4.570107460021973, "identity": 0.7357683777809143, "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_AUTH_reg/fr/c95d74468076fa532b7bb2ea.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "1128a2723671988eeed2a0b03f6966a6264d0756572dba5008a5814007e1df26", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.834733486175537, "target_z": 1.5624998493929763, "ce": 6.0305562019348145, "genuineness": 4.331617832183838, "identity": 0.7644199132919312, "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_AUTH_reg/fr/dadbcf7a3db742ca2d0e6b59.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "d8defc24f41327308b0bcdd1f1e9db9b00ec59b83a3afea34b7eb3e4c5adab90", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.7941524982452393, "target_z": 1.5234374937247073, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.642747163772583, "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_AUTH_reg/fr/5fe19941fbad08f2f4a9135c.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "f2fabb4bbf2226f9cfb41420daea53eaa1d153fa36f3a16d2c7e4c45736107ff", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.6724090576171875, "target_z": 1.4062499677270663, "ce": 6.054384231567383, "genuineness": 4.482243061065674, "identity": 0.5903323888778687, "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_AUTH_reg/fr/02c7768e24a11de718816ca6.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "73af7675db3d86661db6f449ec1f6808af430924f504b6083936d1f534869247", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.181467056274414, "genuineness": 4.444586753845215, "identity": 0.6260895729064941, "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_AUTH_reg/fr/e5db5b8b8f7cbdf5267a390d.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-0-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "369b163afba0a972195ee8bf82f1fc7e5a910e6c9bbc8da1e28e817d79b009eb", "task_id": "vn_S_AUTH_reg-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.141753673553467, "genuineness": 4.293961524963379, "identity": 0.8431411981582642, "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_AUTH_reg/fr/873dcb42992ae938c3300318.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "e5812f811c3bbbb0e0ba3a7df60b9ca2e03d8def7ce09eaa50f059e8a01324d3", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.169203042984009, "target_z": 0.921875059167045, "ce": 6.125868320465088, "genuineness": 4.6705241203308105, "identity": 0.6700038909912109, "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_AUTH_reg/fr/51af20b600f251e4f9ab2c23.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "968e275c13ed2a9827da938ec10941bd8031072426c4ff616970463dec6a20b1", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8834309577941895, "target_z": 1.6093749515905995, "ce": 6.252951145172119, "genuineness": 4.268857479095459, "identity": 0.4464636445045471, "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_AUTH_reg/fr/4db528fd8dfce70da97ec85f.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "1f899cfbb70a580822822ba08e082ae969c53b467c5ba27c17cf53a572d72c24", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.9987621307373047, "target_z": 0.7578124309717807, "ce": 5.990842819213867, "genuineness": 4.469690799713135, "identity": 0.7531707286834717, "hypothesis": "J'ai trouvé la petite bar. J'étais près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "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/audio/vn_S_AUTH_reg/fr/47bcbef4b824fc4216649308.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "7c3a6e6d430c72609e3bf4b504c849391d5f4c9aef308f27f303f46adc383486", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.8671984672546387, "target_z": 1.5937499175247252, "ce": 6.268836498260498, "genuineness": 4.406930446624756, "identity": 0.6599792242050171, "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_AUTH_reg/fr/8954bc26d9e53080315abfc1.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "65f7df91709390009084d5dc28c03392b1eec2da9e16f0f031fda2a1ef3b504c", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.680525302886963, "target_z": 1.4140624847600036, "ce": 6.133810997009277, "genuineness": 4.356721878051758, "identity": 0.7444282174110413, "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_AUTH_reg/fr/77ae6304680e95c74ab9b230.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "7cf10a4837d580974ef7db3be2837c8f6f8a63cf0cd8adea93afb10be01cb4bf", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.1042730808258057, "target_z": 0.8593749229035473, "ce": 6.038498878479004, "genuineness": 4.6705241203308105, "identity": 0.83545982837677, "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_AUTH_reg/fr/303f146ddd368b09f820f1e8.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "aa1438d3e2b62d1f91f6462cd2710ac3dce7f1ecd5dcbd0083cbbdb9d0a7cf52", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 1.9703553915023804, "target_z": 0.7304687361047091, "ce": 6.062326908111572, "genuineness": 4.770940780639648, "identity": 0.8138415217399597, "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_AUTH_reg/fr/0e8bbf216d467466251f1d92.mp3" }, { "key": "original-vn_S_AUTH_reg-fr-1-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "005437f5cc43bfa743e09441b342c67f316e5886383e895f915835ccab7e63e6", "task_id": "vn_S_AUTH_reg-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.6886415481567383, "target_z": 1.4218750017929407, "ce": 6.086154937744141, "genuineness": 4.469690799713135, "identity": 0.8026015758514404, "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_AUTH_reg/fr/840abaf3fb14c37006215b3e.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "70f0c9985d51349517822c1fac4952680aafdc2dd1a455982372f4817bc97c9f", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.2747139930725098, "target_z": 1.0234375510988116, "ce": 6.1020402908325195, "genuineness": 4.20609712600708, "identity": 0.6063812971115112, "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_AUTH_reg/nl/d05a845a6dee1e02f115ddba.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "c49bdf05f4ac2ca5d314798d2c7a181c37420bd7812fba9dbe7d6e6f596bd2b1", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.337939739227295, "target_z": 2.0468749874494145, "ce": 6.133810997009277, "genuineness": 4.319066047668457, "identity": 0.642934262752533, "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_AUTH_reg/nl/52020fa118e2966ac63687e1.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a8b554d8a9dcc95f841111cc360c3913e6602638effda47341b2e5be442018e8", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.6463565826416016, "target_z": 2.343750175708194, "ce": 6.09409761428833, "genuineness": 4.042920112609863, "identity": 0.6680026054382324, "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_AUTH_reg/nl/ce69f50907fb177b038661c0.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "0be14af43cfe1c849f4ea0721000043b2561633eeca091a1bda75ef42b90a10d", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.7373390197753906, "target_z": 1.468750103990564, "ce": 6.109982967376709, "genuineness": 4.419482707977295, "identity": 0.7435582876205444, "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_AUTH_reg/nl/d183b2448cf7dbaafaf5e3e3.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "b3d0e6971b40102197f9075724a349e36181d99f0d5fbfde1368f5404d45a3fd", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.2665977478027344, "target_z": 1.0156250340658743, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.6908595561981201, "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_AUTH_reg/nl/29ed1a25d93a378acef88649.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "c74754ff49b36aea51eaf6d818452bb80ce25c779b54b893ec3b740fed1ca679", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.256777286529541, "target_z": 1.9687498171200426, "ce": 6.141753673553467, "genuineness": 4.331617832183838, "identity": 0.561752200126648, "hypothesis": "De deur staat nog open. Ik heb je jas op de op de stoel 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/audio/vn_S_AUTH_reg/nl/3e54aab055d521aa17248024.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "db11c2ef74b8f3d2330330750bcb99faebd50d7ac9c00a2e43fc5ba1d697f5d1", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.8411459922790527, "target_z": 2.5312501255058533, "ce": 6.117925643920898, "genuineness": 4.2374773025512695, "identity": 0.6466192007064819, "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_AUTH_reg/nl/f80029269a9f5b553f3c7001.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-0-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "056c059eb8d4b27910b0a8800787033e4de783f576beb5c84c0ecae82128d78a", "task_id": "vn_S_AUTH_reg-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.388340950012207, "target_z": 1.1328123305670983, "ce": 6.125868320465088, "genuineness": 4.319066047668457, "identity": 0.6156368255615234, "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_AUTH_reg/nl/390bbfe3e0122281f0a9edeb.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1701-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "86538077f428c4997da9b5f9910a02b8b7374b638a79db512cd5fe2bde851dfd", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.0782203674316406, "target_z": 1.7968749013882583, "ce": 6.24500846862793, "genuineness": 4.218649387359619, "identity": 0.7157139182090759, "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_AUTH_reg/nl/0e9318a5f350046e061f33af.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1701-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "8bf1c3e9002aca9e87bc63b0de3f6ef6fbbdf22eaa966a6d6f8cb117c577e043", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.875314712524414, "target_z": 1.6015624345576622, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.668890655040741, "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_AUTH_reg/nl/6f0c643751343b216a6a9dcc.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1701-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "e0b4a4124f3eef7cdf8ac20102aa23a049463e15790db30ace410bdc0c510a5a", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.126917839050293, "target_z": 1.8437500035858816, "ce": 6.141753673553467, "genuineness": 4.268857479095459, "identity": 0.7459632754325867, "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_AUTH_reg/nl/84d7d167d0d68dbe469d07a7.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1701-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "eace71eac9d12511d7f842e927fce13241a41db72860689eb48d96beb48f86c9", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 4.182027339935303, "target_z": 2.8593749229035472, "ce": 6.213237762451172, "genuineness": 4.093128204345703, "identity": 0.7274882793426514, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar 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/audio/vn_S_AUTH_reg/nl/17d8c010ab40b1cdcf9b1c02.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1702-scale0.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "9a3525fa4ed668239da0905def595ab412559a1e2de6d3ce940067f21b547b21", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.2080800533294678, "target_z": 1.9218749444188366, "ce": 6.292664527893066, "genuineness": 4.356721878051758, "identity": 0.6065389513969421, "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_AUTH_reg/nl/503a21d4d3df85f2b22b1f0c.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1702-scale0.5", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "7cc996792ffb67dd1aeba43209dbcc0bb80f6fa94ea496096adcebf5c80d74fe", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.548961639404297, "target_z": 2.2499999713129477, "ce": 6.086154937744141, "genuineness": 4.20609712600708, "identity": 0.7098926305770874, "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_AUTH_reg/nl/8d807177f4faa56b0822e2ee.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1702-scale0.75", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "17c0f5227a5046bec3a19542d938cda153f8761ab12636cb03d42c68550a5324", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 2.834733486175537, "target_z": 1.5624998493929763, "ce": 6.157639026641846, "genuineness": 4.406930446624756, "identity": 0.6004860997200012, "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_AUTH_reg/nl/2c5ef7cc4bab62983ec1e31b.mp3" }, { "key": "original-vn_S_AUTH_reg-nl-1-seed1702-scale1.0", "bucket": "vn_S_AUTH_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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Strong Authoritative Style. 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": "8a05e9f5b07fbb6fbd545a034e0bc89447086616be62ecf40ecdf2ddb394ccac", "mp3_sha256": "a1d667d3bee474319c02974994b87b58f8486429f27a9cce7a70e88c9c9f28f1", "task_id": "vn_S_AUTH_reg-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Strong Authoritative Style. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Strong Authoritative Style. 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 Authoritative Style. 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 Authoritative Style. 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": 3.419102191925049, "target_z": 2.125000157778787, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.5061518549919128, "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_AUTH_reg/nl/0393823ac65c6d95eb038199.mp3" } ], "source_standard_audio_unchanged": true }